diff --git a/en/python/resources/realtime/index.md b/en/python/resources/realtime/index.md new file mode 100644 index 0000000..50e1ef2 --- /dev/null +++ b/en/python/resources/realtime/index.md @@ -0,0 +1,25577 @@ +# Realtime + +## + +`realtime.connect(RealtimeConnectParams**kwargs)` + +**** `` + +The Realtime API enables you to build low-latency, multi-modal conversational experiences. It currently supports text and audio as both input and output, as well as function calling. + +Some notable benefits of the API include: + +- Native speech-to-speech: Skipping an intermediate text format means low latency and nuanced output. +- Natural, steerable voices: The models have natural inflection and can laugh, whisper, and adhere to tone direction. +- Simultaneous multimodal output: Text is useful for moderation; faster-than-realtime audio ensures stable playback. + +The Realtime API is a stateful, event-based API that communicates over a WebSocket. + +### Parameters + +- `call_id: Optional[str]` + +- `model: Optional[str]` + +### Example + +```python +import os +from openai import OpenAI + +client = OpenAI( + api_key=os.environ.get("OPENAI_API_KEY"), # This is the default and can be omitted +) +client.realtime.connect() +``` + +## Domain Types + +### Audio Transcription + +- `class AudioTranscription: …` + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + +### Conversation Created Event + +- `class ConversationCreatedEvent: …` + + Returned when a conversation is created. Emitted right after session creation. + + - `conversation: Conversation` + + The conversation resource. + + - `id: Optional[str]` + + The unique ID of the conversation. + + - `object: Optional[Literal["realtime.conversation"]]` + + The object type, must be `realtime.conversation`. + + - `"realtime.conversation"` + + - `event_id: str` + + The unique ID of the server event. + + - `type: Literal["conversation.created"]` + + The event type, must be `conversation.created`. + + - `"conversation.created"` + +### Conversation Item + +- `ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + +### Conversation Item Added + +- `class ConversationItemAdded: …` + + Sent by the server when an Item is added to the default Conversation. This can happen in several cases: + + - When the client sends a `conversation.item.create` event. + - When the input audio buffer is committed. In this case the item will be a user message containing the audio from the buffer. + - When the model is generating a Response. In this case the `conversation.item.added` event will be sent when the model starts generating a specific Item, and thus it will not yet have any content (and `status` will be `in_progress`). + + The event will include the full content of the Item (except when model is generating a Response) except for audio data, which can be retrieved separately with a `conversation.item.retrieve` event if necessary. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `type: Literal["conversation.item.added"]` + + The event type, must be `conversation.item.added`. + + - `"conversation.item.added"` + + - `previous_item_id: Optional[str]` + + The ID of the item that precedes this one, if any. This is used to + maintain ordering when items are inserted. + +### Conversation Item Create Event + +- `class ConversationItemCreateEvent: …` + + Add a new Item to the Conversation's context, including messages, function + calls, and function call responses. This event can be used both to populate a + "history" of the conversation and to add new items mid-stream, but has the + current limitation that it cannot populate assistant audio messages. + + If successful, the server will respond with a `conversation.item.created` + event, otherwise an `error` event will be sent. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `type: Literal["conversation.item.create"]` + + The event type, must be `conversation.item.create`. + + - `"conversation.item.create"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `previous_item_id: Optional[str]` + + The ID of the preceding item after which the new item will be inserted. If not set, the new item will be appended to the end of the conversation. + + If set to `root`, the new item will be added to the beginning of the conversation. + + If set to an existing ID, it allows an item to be inserted mid-conversation. If the ID cannot be found, an error will be returned and the item will not be added. + +### Conversation Item Created Event + +- `class ConversationItemCreatedEvent: …` + + Returned when a conversation item is created. There are several scenarios that produce this event: + + - The server is generating a Response, which if successful will produce + either one or two Items, which will be of type `message` + (role `assistant`) or type `function_call`. + - The input audio buffer has been committed, either by the client or the + server (in `server_vad` mode). The server will take the content of the + input audio buffer and add it to a new user message Item. + - The client has sent a `conversation.item.create` event to add a new Item + to the Conversation. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `type: Literal["conversation.item.created"]` + + The event type, must be `conversation.item.created`. + + - `"conversation.item.created"` + + - `previous_item_id: Optional[str]` + + The ID of the preceding item in the Conversation context, allows the + client to understand the order of the conversation. Can be `null` if the + item has no predecessor. + +### Conversation Item Delete Event + +- `class ConversationItemDeleteEvent: …` + + Send this event when you want to remove any item from the conversation + history. The server will respond with a `conversation.item.deleted` event, + unless the item does not exist in the conversation history, in which case the + server will respond with an error. + + - `item_id: str` + + The ID of the item to delete. + + - `type: Literal["conversation.item.delete"]` + + The event type, must be `conversation.item.delete`. + + - `"conversation.item.delete"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + +### Conversation Item Deleted Event + +- `class ConversationItemDeletedEvent: …` + + Returned when an item in the conversation is deleted by the client with a + `conversation.item.delete` event. This event is used to synchronize the + server's understanding of the conversation history with the client's view. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item that was deleted. + + - `type: Literal["conversation.item.deleted"]` + + The event type, must be `conversation.item.deleted`. + + - `"conversation.item.deleted"` + +### Conversation Item Done + +- `class ConversationItemDone: …` + + Returned when a conversation item is finalized. + + The event will include the full content of the Item except for audio data, which can be retrieved separately with a `conversation.item.retrieve` event if needed. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `type: Literal["conversation.item.done"]` + + The event type, must be `conversation.item.done`. + + - `"conversation.item.done"` + + - `previous_item_id: Optional[str]` + + The ID of the item that precedes this one, if any. This is used to + maintain ordering when items are inserted. + +### Conversation Item Input Audio Transcription Completed Event + +- `class ConversationItemInputAudioTranscriptionCompletedEvent: …` + + This event is the output of audio transcription for user audio written to the + user audio buffer. Transcription begins when the input audio buffer is + committed by the client or server (when VAD is enabled). Transcription runs + asynchronously with Response creation, so this event may come before or after + the Response events. + + Realtime API models accept audio natively, and thus input transcription is a + separate process run on a separate ASR (Automatic Speech Recognition) model. + The transcript may diverge somewhat from the model's interpretation, and + should be treated as a rough guide. + + - `content_index: int` + + The index of the content part containing the audio. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item containing the audio that is being transcribed. + + - `transcript: str` + + The transcribed text. + + - `type: Literal["conversation.item.input_audio_transcription.completed"]` + + The event type, must be + `conversation.item.input_audio_transcription.completed`. + + - `"conversation.item.input_audio_transcription.completed"` + + - `usage: Usage` + + Usage statistics for the transcription, this is billed according to the ASR model's pricing rather than the realtime model's pricing. + + - `class UsageTranscriptTextUsageTokens: …` + + Usage statistics for models billed by token usage. + + - `input_tokens: int` + + Number of input tokens billed for this request. + + - `output_tokens: int` + + Number of output tokens generated. + + - `total_tokens: int` + + Total number of tokens used (input + output). + + - `type: Literal["tokens"]` + + The type of the usage object. Always `tokens` for this variant. + + - `"tokens"` + + - `input_token_details: Optional[UsageTranscriptTextUsageTokensInputTokenDetails]` + + Details about the input tokens billed for this request. + + - `audio_tokens: Optional[int]` + + Number of audio tokens billed for this request. + + - `text_tokens: Optional[int]` + + Number of text tokens billed for this request. + + - `class UsageTranscriptTextUsageDuration: …` + + Usage statistics for models billed by audio input duration. + + - `seconds: float` + + Duration of the input audio in seconds. + + - `type: Literal["duration"]` + + The type of the usage object. Always `duration` for this variant. + + - `"duration"` + + - `logprobs: Optional[List[LogProbProperties]]` + + The log probabilities of the transcription. + + - `token: str` + + The token that was used to generate the log probability. + + - `bytes: List[int]` + + The bytes that were used to generate the log probability. + + - `logprob: float` + + The log probability of the token. + +### Conversation Item Input Audio Transcription Delta Event + +- `class ConversationItemInputAudioTranscriptionDeltaEvent: …` + + Returned when the text value of an input audio transcription content part is updated with incremental transcription results. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item containing the audio that is being transcribed. + + - `type: Literal["conversation.item.input_audio_transcription.delta"]` + + The event type, must be `conversation.item.input_audio_transcription.delta`. + + - `"conversation.item.input_audio_transcription.delta"` + + - `content_index: Optional[int]` + + The index of the content part in the item's content array. + + - `delta: Optional[str]` + + The text delta. + + - `logprobs: Optional[List[LogProbProperties]]` + + The log probabilities of the transcription. These can be enabled by configurating the session with `"include": ["item.input_audio_transcription.logprobs"]`. Each entry in the array corresponds a log probability of which token would be selected for this chunk of transcription. This can help to identify if it was possible there were multiple valid options for a given chunk of transcription. + + - `token: str` + + The token that was used to generate the log probability. + + - `bytes: List[int]` + + The bytes that were used to generate the log probability. + + - `logprob: float` + + The log probability of the token. + +### Conversation Item Input Audio Transcription Failed Event + +- `class ConversationItemInputAudioTranscriptionFailedEvent: …` + + Returned when input audio transcription is configured, and a transcription + request for a user message failed. These events are separate from other + `error` events so that the client can identify the related Item. + + - `content_index: int` + + The index of the content part containing the audio. + + - `error: Error` + + Details of the transcription error. + + - `code: Optional[str]` + + Error code, if any. + + - `message: Optional[str]` + + A human-readable error message. + + - `param: Optional[str]` + + Parameter related to the error, if any. + + - `type: Optional[str]` + + The type of error. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the user message item. + + - `type: Literal["conversation.item.input_audio_transcription.failed"]` + + The event type, must be + `conversation.item.input_audio_transcription.failed`. + + - `"conversation.item.input_audio_transcription.failed"` + +### Conversation Item Input Audio Transcription Segment + +- `class ConversationItemInputAudioTranscriptionSegment: …` + + Returned when an input audio transcription segment is identified for an item. + + - `id: str` + + The segment identifier. + + - `content_index: int` + + The index of the input audio content part within the item. + + - `end: float` + + End time of the segment in seconds. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item containing the input audio content. + + - `speaker: str` + + The detected speaker label for this segment. + + - `start: float` + + Start time of the segment in seconds. + + - `text: str` + + The text for this segment. + + - `type: Literal["conversation.item.input_audio_transcription.segment"]` + + The event type, must be `conversation.item.input_audio_transcription.segment`. + + - `"conversation.item.input_audio_transcription.segment"` + +### Conversation Item Retrieve Event + +- `class ConversationItemRetrieveEvent: …` + + Send this event when you want to retrieve the server's representation of a specific item in the conversation history. This is useful, for example, to inspect user audio after noise cancellation and VAD. + The server will respond with a `conversation.item.retrieved` event, + unless the item does not exist in the conversation history, in which case the + server will respond with an error. + + - `item_id: str` + + The ID of the item to retrieve. + + - `type: Literal["conversation.item.retrieve"]` + + The event type, must be `conversation.item.retrieve`. + + - `"conversation.item.retrieve"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + +### Conversation Item Truncate Event + +- `class ConversationItemTruncateEvent: …` + + Send this event to truncate a previous assistant message’s audio. The server + will produce audio faster than realtime, so this event is useful when the user + interrupts to truncate audio that has already been sent to the client but not + yet played. This will synchronize the server's understanding of the audio with + the client's playback. + + Truncating audio will delete the server-side text transcript to ensure there + is not text in the context that hasn't been heard by the user. + + If successful, the server will respond with a `conversation.item.truncated` + event. + + - `audio_end_ms: int` + + Inclusive duration up to which audio is truncated, in milliseconds. If + the audio_end_ms is greater than the actual audio duration, the server + will respond with an error. + + - `content_index: int` + + The index of the content part to truncate. Set this to `0`. + + - `item_id: str` + + The ID of the assistant message item to truncate. Only assistant message + items can be truncated. + + - `type: Literal["conversation.item.truncate"]` + + The event type, must be `conversation.item.truncate`. + + - `"conversation.item.truncate"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + +### Conversation Item Truncated Event + +- `class ConversationItemTruncatedEvent: …` + + Returned when an earlier assistant audio message item is truncated by the + client with a `conversation.item.truncate` event. This event is used to + synchronize the server's understanding of the audio with the client's playback. + + This action will truncate the audio and remove the server-side text transcript + to ensure there is no text in the context that hasn't been heard by the user. + + - `audio_end_ms: int` + + The duration up to which the audio was truncated, in milliseconds. + + - `content_index: int` + + The index of the content part that was truncated. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the assistant message item that was truncated. + + - `type: Literal["conversation.item.truncated"]` + + The event type, must be `conversation.item.truncated`. + + - `"conversation.item.truncated"` + +### Conversation Item With Reference + +- `class ConversationItemWithReference: …` + + The item to add to the conversation. + + - `id: Optional[str]` + + For an item of type (`message` | `function_call` | `function_call_output`) + this field allows the client to assign the unique ID of the item. It is + not required because the server will generate one if not provided. + + For an item of type `item_reference`, this field is required and is a + reference to any item that has previously existed in the conversation. + + - `arguments: Optional[str]` + + The arguments of the function call (for `function_call` items). + + - `call_id: Optional[str]` + + The ID of the function call (for `function_call` and + `function_call_output` items). If passed on a `function_call_output` + item, the server will check that a `function_call` item with the same + ID exists in the conversation history. + + - `content: Optional[List[Content]]` + + The content of the message, applicable for `message` items. + + - Message items of role `system` support only `input_text` content + - Message items of role `user` support `input_text` and `input_audio` + content + - Message items of role `assistant` support `text` content. + + - `id: Optional[str]` + + ID of a previous conversation item to reference (for `item_reference` + content types in `response.create` events). These can reference both + client and server created items. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, used for `input_audio` content type. + + - `text: Optional[str]` + + The text content, used for `input_text` and `text` content types. + + - `transcript: Optional[str]` + + The transcript of the audio, used for `input_audio` content type. + + - `type: Optional[Literal["input_text", "input_audio", "item_reference", "text"]]` + + The content type (`input_text`, `input_audio`, `item_reference`, `text`). + + - `"input_text"` + + - `"input_audio"` + + - `"item_reference"` + + - `"text"` + + - `name: Optional[str]` + + The name of the function being called (for `function_call` items). + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. + + - `"realtime.item"` + + - `output: Optional[str]` + + The output of the function call (for `function_call_output` items). + + - `role: Optional[Literal["user", "assistant", "system"]]` + + The role of the message sender (`user`, `assistant`, `system`), only + applicable for `message` items. + + - `"user"` + + - `"assistant"` + + - `"system"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item (`completed`, `incomplete`, `in_progress`). These have no effect + on the conversation, but are accepted for consistency with the + `conversation.item.created` event. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `type: Optional[Literal["message", "function_call", "function_call_output", "item_reference"]]` + + The type of the item (`message`, `function_call`, `function_call_output`, `item_reference`). + + - `"message"` + + - `"function_call"` + + - `"function_call_output"` + + - `"item_reference"` + +### Input Audio Buffer Append Event + +- `class InputAudioBufferAppendEvent: …` + + Send this event to append audio bytes to the input audio buffer. The audio + buffer is temporary storage you can write to and later commit. A "commit" will create a new + user message item in the conversation history from the buffer content and clear the buffer. + Input audio transcription (if enabled) will be generated when the buffer is committed. + + If VAD is enabled the audio buffer is used to detect speech and the server will decide + when to commit. When Server VAD is disabled, you must commit the audio buffer + manually. Input audio noise reduction operates on writes to the audio buffer. + + The client may choose how much audio to place in each event up to a maximum + of 15 MiB, for example streaming smaller chunks from the client may allow the + VAD to be more responsive. Unlike most other client events, the server will + not send a confirmation response to this event. + + - `audio: str` + + Base64-encoded audio bytes. This must be in the format specified by the + `input_audio_format` field in the session configuration. + + - `type: Literal["input_audio_buffer.append"]` + + The event type, must be `input_audio_buffer.append`. + + - `"input_audio_buffer.append"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + +### Input Audio Buffer Clear Event + +- `class InputAudioBufferClearEvent: …` + + Send this event to clear the audio bytes in the buffer. The server will + respond with an `input_audio_buffer.cleared` event. + + - `type: Literal["input_audio_buffer.clear"]` + + The event type, must be `input_audio_buffer.clear`. + + - `"input_audio_buffer.clear"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + +### Input Audio Buffer Cleared Event + +- `class InputAudioBufferClearedEvent: …` + + Returned when the input audio buffer is cleared by the client with a + `input_audio_buffer.clear` event. + + - `event_id: str` + + The unique ID of the server event. + + - `type: Literal["input_audio_buffer.cleared"]` + + The event type, must be `input_audio_buffer.cleared`. + + - `"input_audio_buffer.cleared"` + +### Input Audio Buffer Commit Event + +- `class InputAudioBufferCommitEvent: …` + + Send this event to commit the user input audio buffer, which will create a new user message item in the conversation. This event will produce an error if the input audio buffer is empty. When in Server VAD mode, the client does not need to send this event, the server will commit the audio buffer automatically. + + Committing the input audio buffer will trigger input audio transcription (if enabled in session configuration), but it will not create a response from the model. The server will respond with an `input_audio_buffer.committed` event. + + - `type: Literal["input_audio_buffer.commit"]` + + The event type, must be `input_audio_buffer.commit`. + + - `"input_audio_buffer.commit"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + +### Input Audio Buffer Committed Event + +- `class InputAudioBufferCommittedEvent: …` + + Returned when an input audio buffer is committed, either by the client or + automatically in server VAD mode. The `item_id` property is the ID of the user + message item that will be created, thus a `conversation.item.created` event + will also be sent to the client. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the user message item that will be created. + + - `type: Literal["input_audio_buffer.committed"]` + + The event type, must be `input_audio_buffer.committed`. + + - `"input_audio_buffer.committed"` + + - `previous_item_id: Optional[str]` + + The ID of the preceding item after which the new item will be inserted. + Can be `null` if the item has no predecessor. + +### Input Audio Buffer Dtmf Event Received Event + +- `class InputAudioBufferDtmfEventReceivedEvent: …` + + **SIP Only:** Returned when an DTMF event is received. A DTMF event is a message that + represents a telephone keypad press (0–9, *, #, A–D). The `event` property + is the keypad that the user press. The `received_at` is the UTC Unix Timestamp + that the server received the event. + + - `event: str` + + The telephone keypad that was pressed by the user. + + - `received_at: int` + + UTC Unix Timestamp when DTMF Event was received by server. + + - `type: Literal["input_audio_buffer.dtmf_event_received"]` + + The event type, must be `input_audio_buffer.dtmf_event_received`. + + - `"input_audio_buffer.dtmf_event_received"` + +### Input Audio Buffer Speech Started Event + +- `class InputAudioBufferSpeechStartedEvent: …` + + Sent by the server when in `server_vad` mode to indicate that speech has been + detected in the audio buffer. This can happen any time audio is added to the + buffer (unless speech is already detected). The client may want to use this + event to interrupt audio playback or provide visual feedback to the user. + + The client should expect to receive a `input_audio_buffer.speech_stopped` event + when speech stops. The `item_id` property is the ID of the user message item + that will be created when speech stops and will also be included in the + `input_audio_buffer.speech_stopped` event (unless the client manually commits + the audio buffer during VAD activation). + + - `audio_start_ms: int` + + Milliseconds from the start of all audio written to the buffer during the + session when speech was first detected. This will correspond to the + beginning of audio sent to the model, and thus includes the + `prefix_padding_ms` configured in the Session. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the user message item that will be created when speech stops. + + - `type: Literal["input_audio_buffer.speech_started"]` + + The event type, must be `input_audio_buffer.speech_started`. + + - `"input_audio_buffer.speech_started"` + +### Input Audio Buffer Speech Stopped Event + +- `class InputAudioBufferSpeechStoppedEvent: …` + + Returned in `server_vad` mode when the server detects the end of speech in + the audio buffer. The server will also send an `conversation.item.created` + event with the user message item that is created from the audio buffer. + + - `audio_end_ms: int` + + Milliseconds since the session started when speech stopped. This will + correspond to the end of audio sent to the model, and thus includes the + `min_silence_duration_ms` configured in the Session. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the user message item that will be created. + + - `type: Literal["input_audio_buffer.speech_stopped"]` + + The event type, must be `input_audio_buffer.speech_stopped`. + + - `"input_audio_buffer.speech_stopped"` + +### Input Audio Buffer Timeout Triggered + +- `class InputAudioBufferTimeoutTriggered: …` + + Returned when the Server VAD timeout is triggered for the input audio buffer. This is configured + with `idle_timeout_ms` in the `turn_detection` settings of the session, and it indicates that + there hasn't been any speech detected for the configured duration. + + The `audio_start_ms` and `audio_end_ms` fields indicate the segment of audio after the last + model response up to the triggering time, as an offset from the beginning of audio written + to the input audio buffer. This means it demarcates the segment of audio that was silent and + the difference between the start and end values will roughly match the configured timeout. + + The empty audio will be committed to the conversation as an `input_audio` item (there will be a + `input_audio_buffer.committed` event) and a model response will be generated. There may be speech + that didn't trigger VAD but is still detected by the model, so the model may respond with + something relevant to the conversation or a prompt to continue speaking. + + - `audio_end_ms: int` + + Millisecond offset of audio written to the input audio buffer at the time the timeout was triggered. + + - `audio_start_ms: int` + + Millisecond offset of audio written to the input audio buffer that was after the playback time of the last model response. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item associated with this segment. + + - `type: Literal["input_audio_buffer.timeout_triggered"]` + + The event type, must be `input_audio_buffer.timeout_triggered`. + + - `"input_audio_buffer.timeout_triggered"` + +### Log Prob Properties + +- `class LogProbProperties: …` + + A log probability object. + + - `token: str` + + The token that was used to generate the log probability. + + - `bytes: List[int]` + + The bytes that were used to generate the log probability. + + - `logprob: float` + + The log probability of the token. + +### Mcp List Tools Completed + +- `class McpListToolsCompleted: …` + + Returned when listing MCP tools has completed for an item. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP list tools item. + + - `type: Literal["mcp_list_tools.completed"]` + + The event type, must be `mcp_list_tools.completed`. + + - `"mcp_list_tools.completed"` + +### Mcp List Tools Failed + +- `class McpListToolsFailed: …` + + Returned when listing MCP tools has failed for an item. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP list tools item. + + - `type: Literal["mcp_list_tools.failed"]` + + The event type, must be `mcp_list_tools.failed`. + + - `"mcp_list_tools.failed"` + +### Mcp List Tools In Progress + +- `class McpListToolsInProgress: …` + + Returned when listing MCP tools is in progress for an item. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP list tools item. + + - `type: Literal["mcp_list_tools.in_progress"]` + + The event type, must be `mcp_list_tools.in_progress`. + + - `"mcp_list_tools.in_progress"` + +### Noise Reduction Type + +- `Literal["near_field", "far_field"]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + +### Output Audio Buffer Clear Event + +- `class OutputAudioBufferClearEvent: …` + + **WebRTC/SIP Only:** Emit to cut off the current audio response. This will trigger the server to + stop generating audio and emit a `output_audio_buffer.cleared` event. This + event should be preceded by a `response.cancel` client event to stop the + generation of the current response. + [Learn more](https://platform.openai.com/docs/guides/realtime-conversations#client-and-server-events-for-audio-in-webrtc). + + - `type: Literal["output_audio_buffer.clear"]` + + The event type, must be `output_audio_buffer.clear`. + + - `"output_audio_buffer.clear"` + + - `event_id: Optional[str]` + + The unique ID of the client event used for error handling. + +### Rate Limits Updated Event + +- `class RateLimitsUpdatedEvent: …` + + Emitted at the beginning of a Response to indicate the updated rate limits. + When a Response is created some tokens will be "reserved" for the output + tokens, the rate limits shown here reflect that reservation, which is then + adjusted accordingly once the Response is completed. + + - `event_id: str` + + The unique ID of the server event. + + - `rate_limits: List[RateLimit]` + + List of rate limit information. + + - `limit: Optional[int]` + + The maximum allowed value for the rate limit. + + - `name: Optional[Literal["requests", "tokens"]]` + + The name of the rate limit (`requests`, `tokens`). + + - `"requests"` + + - `"tokens"` + + - `remaining: Optional[int]` + + The remaining value before the limit is reached. + + - `reset_seconds: Optional[float]` + + Seconds until the rate limit resets. + + - `type: Literal["rate_limits.updated"]` + + The event type, must be `rate_limits.updated`. + + - `"rate_limits.updated"` + +### Realtime Audio Config + +- `class RealtimeAudioConfig: …` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + +### Realtime Audio Config Input + +- `class RealtimeAudioConfigInput: …` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + +### Realtime Audio Config Output + +- `class RealtimeAudioConfigOutput: …` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + +### Realtime Audio Formats + +- `RealtimeAudioFormats` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + +### Realtime Audio Input Turn Detection + +- `Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + +### Realtime Client Event + +- `RealtimeClientEvent` + + A realtime client event. + + - `class ConversationItemCreateEvent: …` + + Add a new Item to the Conversation's context, including messages, function + calls, and function call responses. This event can be used both to populate a + "history" of the conversation and to add new items mid-stream, but has the + current limitation that it cannot populate assistant audio messages. + + If successful, the server will respond with a `conversation.item.created` + event, otherwise an `error` event will be sent. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `type: Literal["conversation.item.create"]` + + The event type, must be `conversation.item.create`. + + - `"conversation.item.create"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `previous_item_id: Optional[str]` + + The ID of the preceding item after which the new item will be inserted. If not set, the new item will be appended to the end of the conversation. + + If set to `root`, the new item will be added to the beginning of the conversation. + + If set to an existing ID, it allows an item to be inserted mid-conversation. If the ID cannot be found, an error will be returned and the item will not be added. + + - `class ConversationItemDeleteEvent: …` + + Send this event when you want to remove any item from the conversation + history. The server will respond with a `conversation.item.deleted` event, + unless the item does not exist in the conversation history, in which case the + server will respond with an error. + + - `item_id: str` + + The ID of the item to delete. + + - `type: Literal["conversation.item.delete"]` + + The event type, must be `conversation.item.delete`. + + - `"conversation.item.delete"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `class ConversationItemRetrieveEvent: …` + + Send this event when you want to retrieve the server's representation of a specific item in the conversation history. This is useful, for example, to inspect user audio after noise cancellation and VAD. + The server will respond with a `conversation.item.retrieved` event, + unless the item does not exist in the conversation history, in which case the + server will respond with an error. + + - `item_id: str` + + The ID of the item to retrieve. + + - `type: Literal["conversation.item.retrieve"]` + + The event type, must be `conversation.item.retrieve`. + + - `"conversation.item.retrieve"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `class ConversationItemTruncateEvent: …` + + Send this event to truncate a previous assistant message’s audio. The server + will produce audio faster than realtime, so this event is useful when the user + interrupts to truncate audio that has already been sent to the client but not + yet played. This will synchronize the server's understanding of the audio with + the client's playback. + + Truncating audio will delete the server-side text transcript to ensure there + is not text in the context that hasn't been heard by the user. + + If successful, the server will respond with a `conversation.item.truncated` + event. + + - `audio_end_ms: int` + + Inclusive duration up to which audio is truncated, in milliseconds. If + the audio_end_ms is greater than the actual audio duration, the server + will respond with an error. + + - `content_index: int` + + The index of the content part to truncate. Set this to `0`. + + - `item_id: str` + + The ID of the assistant message item to truncate. Only assistant message + items can be truncated. + + - `type: Literal["conversation.item.truncate"]` + + The event type, must be `conversation.item.truncate`. + + - `"conversation.item.truncate"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `class InputAudioBufferAppendEvent: …` + + Send this event to append audio bytes to the input audio buffer. The audio + buffer is temporary storage you can write to and later commit. A "commit" will create a new + user message item in the conversation history from the buffer content and clear the buffer. + Input audio transcription (if enabled) will be generated when the buffer is committed. + + If VAD is enabled the audio buffer is used to detect speech and the server will decide + when to commit. When Server VAD is disabled, you must commit the audio buffer + manually. Input audio noise reduction operates on writes to the audio buffer. + + The client may choose how much audio to place in each event up to a maximum + of 15 MiB, for example streaming smaller chunks from the client may allow the + VAD to be more responsive. Unlike most other client events, the server will + not send a confirmation response to this event. + + - `audio: str` + + Base64-encoded audio bytes. This must be in the format specified by the + `input_audio_format` field in the session configuration. + + - `type: Literal["input_audio_buffer.append"]` + + The event type, must be `input_audio_buffer.append`. + + - `"input_audio_buffer.append"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `class InputAudioBufferClearEvent: …` + + Send this event to clear the audio bytes in the buffer. The server will + respond with an `input_audio_buffer.cleared` event. + + - `type: Literal["input_audio_buffer.clear"]` + + The event type, must be `input_audio_buffer.clear`. + + - `"input_audio_buffer.clear"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `class OutputAudioBufferClearEvent: …` + + **WebRTC/SIP Only:** Emit to cut off the current audio response. This will trigger the server to + stop generating audio and emit a `output_audio_buffer.cleared` event. This + event should be preceded by a `response.cancel` client event to stop the + generation of the current response. + [Learn more](https://platform.openai.com/docs/guides/realtime-conversations#client-and-server-events-for-audio-in-webrtc). + + - `type: Literal["output_audio_buffer.clear"]` + + The event type, must be `output_audio_buffer.clear`. + + - `"output_audio_buffer.clear"` + + - `event_id: Optional[str]` + + The unique ID of the client event used for error handling. + + - `class InputAudioBufferCommitEvent: …` + + Send this event to commit the user input audio buffer, which will create a new user message item in the conversation. This event will produce an error if the input audio buffer is empty. When in Server VAD mode, the client does not need to send this event, the server will commit the audio buffer automatically. + + Committing the input audio buffer will trigger input audio transcription (if enabled in session configuration), but it will not create a response from the model. The server will respond with an `input_audio_buffer.committed` event. + + - `type: Literal["input_audio_buffer.commit"]` + + The event type, must be `input_audio_buffer.commit`. + + - `"input_audio_buffer.commit"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `class ResponseCancelEvent: …` + + Send this event to cancel an in-progress response. The server will respond + with a `response.done` event with a status of `response.status=cancelled`. If + there is no response to cancel, the server will respond with an error. It's safe + to call `response.cancel` even if no response is in progress, an error will be + returned the session will remain unaffected. + + - `type: Literal["response.cancel"]` + + The event type, must be `response.cancel`. + + - `"response.cancel"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `response_id: Optional[str]` + + A specific response ID to cancel - if not provided, will cancel an + in-progress response in the default conversation. + + - `class ResponseCreateEvent: …` + + This event instructs the server to create a Response, which means triggering + model inference. When in Server VAD mode, the server will create Responses + automatically. + + A Response will include at least one Item, and may have two, in which case + the second will be a function call. These Items will be appended to the + conversation history by default. + + The server will respond with a `response.created` event, events for Items + and content created, and finally a `response.done` event to indicate the + Response is complete. + + The `response.create` event includes inference configuration like + `instructions` and `tools`. If these are set, they will override the Session's + configuration for this Response only. + + Responses can be created out-of-band of the default Conversation, meaning that they can + have arbitrary input, and it's possible to disable writing the output to the Conversation. + Only one Response can write to the default Conversation at a time, but otherwise multiple + Responses can be created in parallel. The `metadata` field is a good way to disambiguate + multiple simultaneous Responses. + + Clients can set `conversation` to `none` to create a Response that does not write to the default + Conversation. Arbitrary input can be provided with the `input` field, which is an array accepting + raw Items and references to existing Items. + + - `type: Literal["response.create"]` + + The event type, must be `response.create`. + + - `"response.create"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `response: Optional[RealtimeResponseCreateParams]` + + Create a new Realtime response with these parameters + + - `audio: Optional[RealtimeResponseCreateAudioOutput]` + + Configuration for audio input and output. + + - `output: Optional[Output]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `voice: Optional[OutputVoice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class OutputVoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `conversation: Optional[Union[str, Literal["auto", "none"], null]]` + + Controls which conversation the response is added to. Currently supports + `auto` and `none`, with `auto` as the default value. The `auto` value + means that the contents of the response will be added to the default + conversation. Set this to `none` to create an out-of-band response which + will not add items to default conversation. + + - `str` + + - `Literal["auto", "none"]` + + Controls which conversation the response is added to. Currently supports + `auto` and `none`, with `auto` as the default value. The `auto` value + means that the contents of the response will be added to the default + conversation. Set this to `none` to create an out-of-band response which + will not add items to default conversation. + + - `"auto"` + + - `"none"` + + - `input: Optional[List[ConversationItem]]` + + Input items to include in the prompt for the model. Using this field + creates a new context for this Response instead of using the default + conversation. An empty array `[]` will clear the context for this Response. + Note that this can include references to items that previously appeared in the session + using their id. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `metadata: Optional[Metadata]` + + Set of 16 key-value pairs that can be attached to an object. This can be + useful for storing additional information about the object in a structured + format, and querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings + with a maximum length of 512 characters. + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model used to respond, currently the only possible values are + `[\"audio\"]`, `[\"text\"]`. Audio output always include a text transcript. Setting the + output to mode `text` will disable audio output from the model. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[ToolChoice]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[List[Tool]]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class RealtimeResponseCreateMcpTool: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[AllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class AllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[RequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class RequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[RequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[RequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `class SessionUpdateEvent: …` + + Send this event to update the session’s configuration. + The client may send this event at any time to update any field + except for `voice` and `model`. `voice` can be updated only if there have been no other audio outputs yet. + + When the server receives a `session.update`, it will respond + with a `session.updated` event showing the full, effective configuration. + Only the fields that are present in the `session.update` are updated. To clear a field like + `instructions`, pass an empty string. To clear a field like `tools`, pass an empty array. + To clear a field like `turn_detection`, pass `null`. + + - `session: Session` + + Update the Realtime session. Choose either a realtime + session or a transcription session. + + - `class RealtimeSessionCreateRequest: …` + + Realtime session object configuration. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[RealtimeAudioConfig]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `tool_choice: Optional[RealtimeToolChoiceConfig]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `tools: Optional[RealtimeToolsConfig]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + + - `class RealtimeTranscriptionSessionCreateRequest: …` + + Realtime transcription session object configuration. + + - `type: Literal["transcription"]` + + The type of session to create. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[RealtimeTranscriptionSessionAudio]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeTranscriptionSessionAudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `type: Literal["session.update"]` + + The event type, must be `session.update`. + + - `"session.update"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. This is an arbitrary string that a client may assign. It will be passed back if there is an error with the event, but the corresponding `session.updated` event will not include it. + +### Realtime Conversation Item Assistant Message + +- `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + +### Realtime Conversation Item Function Call + +- `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + +### Realtime Conversation Item Function Call Output + +- `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + +### Realtime Conversation Item System Message + +- `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + +### Realtime Conversation Item User Message + +- `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + +### Realtime Error + +- `class RealtimeError: …` + + Details of the error. + + - `message: str` + + A human-readable error message. + + - `type: str` + + The type of error (e.g., "invalid_request_error", "server_error"). + + - `code: Optional[str]` + + Error code, if any. + + - `event_id: Optional[str]` + + The event_id of the client event that caused the error, if applicable. + + - `param: Optional[str]` + + Parameter related to the error, if any. + +### Realtime Error Event + +- `class RealtimeErrorEvent: …` + + Returned when an error occurs, which could be a client problem or a server + problem. Most errors are recoverable and the session will stay open, we + recommend to implementors to monitor and log error messages by default. + + - `error: RealtimeError` + + Details of the error. + + - `message: str` + + A human-readable error message. + + - `type: str` + + The type of error (e.g., "invalid_request_error", "server_error"). + + - `code: Optional[str]` + + Error code, if any. + + - `event_id: Optional[str]` + + The event_id of the client event that caused the error, if applicable. + + - `param: Optional[str]` + + Parameter related to the error, if any. + + - `event_id: str` + + The unique ID of the server event. + + - `type: Literal["error"]` + + The event type, must be `error`. + + - `"error"` + +### Realtime Function Tool + +- `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + +### Realtime Mcp Approval Request + +- `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + +### Realtime Mcp Approval Response + +- `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + +### Realtime Mcp List Tools + +- `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + +### Realtime Mcp Protocol Error + +- `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + +### Realtime Mcp Tool Call + +- `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + +### Realtime Mcp Tool Execution Error + +- `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + +### Realtime Mcphttp Error + +- `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + +### Realtime Response + +- `class RealtimeResponse: …` + + The response resource. + + - `id: Optional[str]` + + The unique ID of the response, will look like `resp_1234`. + + - `audio: Optional[Audio]` + + Configuration for audio output. + + - `output: Optional[AudioOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `voice: Optional[Union[str, Literal["alloy", "ash", "ballad", 7 more], null]]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `conversation_id: Optional[str]` + + Which conversation the response is added to, determined by the `conversation` + field in the `response.create` event. If `auto`, the response will be added to + the default conversation and the value of `conversation_id` will be an id like + `conv_1234`. If `none`, the response will not be added to any conversation and + the value of `conversation_id` will be `null`. If responses are being triggered + automatically by VAD the response will be added to the default conversation + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls, that was used in this response. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `metadata: Optional[Metadata]` + + Set of 16 key-value pairs that can be attached to an object. This can be + useful for storing additional information about the object in a structured + format, and querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings + with a maximum length of 512 characters. + + - `object: Optional[Literal["realtime.response"]]` + + The object type, must be `realtime.response`. + + - `"realtime.response"` + + - `output: Optional[List[ConversationItem]]` + + The list of output items generated by the response. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model used to respond, currently the only possible values are + `[\"audio\"]`, `[\"text\"]`. Audio output always include a text transcript. Setting the + output to mode `text` will disable audio output from the model. + + - `"text"` + + - `"audio"` + + - `status: Optional[Literal["completed", "cancelled", "failed", 2 more]]` + + The final status of the response (`completed`, `cancelled`, `failed`, or + `incomplete`, `in_progress`). + + - `"completed"` + + - `"cancelled"` + + - `"failed"` + + - `"incomplete"` + + - `"in_progress"` + + - `status_details: Optional[RealtimeResponseStatus]` + + Additional details about the status. + + - `error: Optional[Error]` + + A description of the error that caused the response to fail, + populated when the `status` is `failed`. + + - `code: Optional[str]` + + Error code, if any. + + - `type: Optional[str]` + + The type of error. + + - `reason: Optional[Literal["turn_detected", "client_cancelled", "max_output_tokens", "content_filter"]]` + + The reason the Response did not complete. For a `cancelled` Response, one of `turn_detected` (the server VAD detected a new start of speech) or `client_cancelled` (the client sent a cancel event). For an `incomplete` Response, one of `max_output_tokens` or `content_filter` (the server-side safety filter activated and cut off the response). + + - `"turn_detected"` + + - `"client_cancelled"` + + - `"max_output_tokens"` + + - `"content_filter"` + + - `type: Optional[Literal["completed", "cancelled", "incomplete", "failed"]]` + + The type of error that caused the response to fail, corresponding + with the `status` field (`completed`, `cancelled`, `incomplete`, + `failed`). + + - `"completed"` + + - `"cancelled"` + + - `"incomplete"` + + - `"failed"` + + - `usage: Optional[RealtimeResponseUsage]` + + Usage statistics for the Response, this will correspond to billing. A + Realtime API session will maintain a conversation context and append new + Items to the Conversation, thus output from previous turns (text and + audio tokens) will become the input for later turns. + + - `input_token_details: Optional[RealtimeResponseUsageInputTokenDetails]` + + Details about the input tokens used in the Response. Cached tokens are tokens from previous turns in the conversation that are included as context for the current response. Cached tokens here are counted as a subset of input tokens, meaning input tokens will include cached and uncached tokens. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used as input for the Response. + + - `cached_tokens: Optional[int]` + + The number of cached tokens used as input for the Response. + + - `cached_tokens_details: Optional[CachedTokensDetails]` + + Details about the cached tokens used as input for the Response. + + - `audio_tokens: Optional[int]` + + The number of cached audio tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of cached image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of cached text tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used as input for the Response. + + - `input_tokens: Optional[int]` + + The number of input tokens used in the Response, including text and + audio tokens. + + - `output_token_details: Optional[RealtimeResponseUsageOutputTokenDetails]` + + Details about the output tokens used in the Response. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used in the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used in the Response. + + - `output_tokens: Optional[int]` + + The number of output tokens sent in the Response, including text and + audio tokens. + + - `total_tokens: Optional[int]` + + The total number of tokens in the Response including input and output + text and audio tokens. + +### Realtime Response Create Audio Output + +- `class RealtimeResponseCreateAudioOutput: …` + + Configuration for audio input and output. + + - `output: Optional[Output]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `voice: Optional[OutputVoice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class OutputVoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + +### Realtime Response Create Mcp Tool + +- `class RealtimeResponseCreateMcpTool: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[AllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class AllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[RequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class RequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[RequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[RequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + +### Realtime Response Create Params + +- `class RealtimeResponseCreateParams: …` + + Create a new Realtime response with these parameters + + - `audio: Optional[RealtimeResponseCreateAudioOutput]` + + Configuration for audio input and output. + + - `output: Optional[Output]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `voice: Optional[OutputVoice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class OutputVoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `conversation: Optional[Union[str, Literal["auto", "none"], null]]` + + Controls which conversation the response is added to. Currently supports + `auto` and `none`, with `auto` as the default value. The `auto` value + means that the contents of the response will be added to the default + conversation. Set this to `none` to create an out-of-band response which + will not add items to default conversation. + + - `str` + + - `Literal["auto", "none"]` + + Controls which conversation the response is added to. Currently supports + `auto` and `none`, with `auto` as the default value. The `auto` value + means that the contents of the response will be added to the default + conversation. Set this to `none` to create an out-of-band response which + will not add items to default conversation. + + - `"auto"` + + - `"none"` + + - `input: Optional[List[ConversationItem]]` + + Input items to include in the prompt for the model. Using this field + creates a new context for this Response instead of using the default + conversation. An empty array `[]` will clear the context for this Response. + Note that this can include references to items that previously appeared in the session + using their id. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `metadata: Optional[Metadata]` + + Set of 16 key-value pairs that can be attached to an object. This can be + useful for storing additional information about the object in a structured + format, and querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings + with a maximum length of 512 characters. + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model used to respond, currently the only possible values are + `[\"audio\"]`, `[\"text\"]`. Audio output always include a text transcript. Setting the + output to mode `text` will disable audio output from the model. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[ToolChoice]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[List[Tool]]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class RealtimeResponseCreateMcpTool: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[AllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class AllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[RequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class RequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[RequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[RequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + +### Realtime Response Status + +- `class RealtimeResponseStatus: …` + + Additional details about the status. + + - `error: Optional[Error]` + + A description of the error that caused the response to fail, + populated when the `status` is `failed`. + + - `code: Optional[str]` + + Error code, if any. + + - `type: Optional[str]` + + The type of error. + + - `reason: Optional[Literal["turn_detected", "client_cancelled", "max_output_tokens", "content_filter"]]` + + The reason the Response did not complete. For a `cancelled` Response, one of `turn_detected` (the server VAD detected a new start of speech) or `client_cancelled` (the client sent a cancel event). For an `incomplete` Response, one of `max_output_tokens` or `content_filter` (the server-side safety filter activated and cut off the response). + + - `"turn_detected"` + + - `"client_cancelled"` + + - `"max_output_tokens"` + + - `"content_filter"` + + - `type: Optional[Literal["completed", "cancelled", "incomplete", "failed"]]` + + The type of error that caused the response to fail, corresponding + with the `status` field (`completed`, `cancelled`, `incomplete`, + `failed`). + + - `"completed"` + + - `"cancelled"` + + - `"incomplete"` + + - `"failed"` + +### Realtime Response Usage + +- `class RealtimeResponseUsage: …` + + Usage statistics for the Response, this will correspond to billing. A + Realtime API session will maintain a conversation context and append new + Items to the Conversation, thus output from previous turns (text and + audio tokens) will become the input for later turns. + + - `input_token_details: Optional[RealtimeResponseUsageInputTokenDetails]` + + Details about the input tokens used in the Response. Cached tokens are tokens from previous turns in the conversation that are included as context for the current response. Cached tokens here are counted as a subset of input tokens, meaning input tokens will include cached and uncached tokens. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used as input for the Response. + + - `cached_tokens: Optional[int]` + + The number of cached tokens used as input for the Response. + + - `cached_tokens_details: Optional[CachedTokensDetails]` + + Details about the cached tokens used as input for the Response. + + - `audio_tokens: Optional[int]` + + The number of cached audio tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of cached image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of cached text tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used as input for the Response. + + - `input_tokens: Optional[int]` + + The number of input tokens used in the Response, including text and + audio tokens. + + - `output_token_details: Optional[RealtimeResponseUsageOutputTokenDetails]` + + Details about the output tokens used in the Response. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used in the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used in the Response. + + - `output_tokens: Optional[int]` + + The number of output tokens sent in the Response, including text and + audio tokens. + + - `total_tokens: Optional[int]` + + The total number of tokens in the Response including input and output + text and audio tokens. + +### Realtime Response Usage Input Token Details + +- `class RealtimeResponseUsageInputTokenDetails: …` + + Details about the input tokens used in the Response. Cached tokens are tokens from previous turns in the conversation that are included as context for the current response. Cached tokens here are counted as a subset of input tokens, meaning input tokens will include cached and uncached tokens. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used as input for the Response. + + - `cached_tokens: Optional[int]` + + The number of cached tokens used as input for the Response. + + - `cached_tokens_details: Optional[CachedTokensDetails]` + + Details about the cached tokens used as input for the Response. + + - `audio_tokens: Optional[int]` + + The number of cached audio tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of cached image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of cached text tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used as input for the Response. + +### Realtime Response Usage Output Token Details + +- `class RealtimeResponseUsageOutputTokenDetails: …` + + Details about the output tokens used in the Response. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used in the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used in the Response. + +### Realtime Server Event + +- `RealtimeServerEvent` + + A realtime server event. + + - `class ConversationCreatedEvent: …` + + Returned when a conversation is created. Emitted right after session creation. + + - `conversation: Conversation` + + The conversation resource. + + - `id: Optional[str]` + + The unique ID of the conversation. + + - `object: Optional[Literal["realtime.conversation"]]` + + The object type, must be `realtime.conversation`. + + - `"realtime.conversation"` + + - `event_id: str` + + The unique ID of the server event. + + - `type: Literal["conversation.created"]` + + The event type, must be `conversation.created`. + + - `"conversation.created"` + + - `class ConversationItemCreatedEvent: …` + + Returned when a conversation item is created. There are several scenarios that produce this event: + + - The server is generating a Response, which if successful will produce + either one or two Items, which will be of type `message` + (role `assistant`) or type `function_call`. + - The input audio buffer has been committed, either by the client or the + server (in `server_vad` mode). The server will take the content of the + input audio buffer and add it to a new user message Item. + - The client has sent a `conversation.item.create` event to add a new Item + to the Conversation. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `type: Literal["conversation.item.created"]` + + The event type, must be `conversation.item.created`. + + - `"conversation.item.created"` + + - `previous_item_id: Optional[str]` + + The ID of the preceding item in the Conversation context, allows the + client to understand the order of the conversation. Can be `null` if the + item has no predecessor. + + - `class ConversationItemDeletedEvent: …` + + Returned when an item in the conversation is deleted by the client with a + `conversation.item.delete` event. This event is used to synchronize the + server's understanding of the conversation history with the client's view. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item that was deleted. + + - `type: Literal["conversation.item.deleted"]` + + The event type, must be `conversation.item.deleted`. + + - `"conversation.item.deleted"` + + - `class ConversationItemInputAudioTranscriptionCompletedEvent: …` + + This event is the output of audio transcription for user audio written to the + user audio buffer. Transcription begins when the input audio buffer is + committed by the client or server (when VAD is enabled). Transcription runs + asynchronously with Response creation, so this event may come before or after + the Response events. + + Realtime API models accept audio natively, and thus input transcription is a + separate process run on a separate ASR (Automatic Speech Recognition) model. + The transcript may diverge somewhat from the model's interpretation, and + should be treated as a rough guide. + + - `content_index: int` + + The index of the content part containing the audio. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item containing the audio that is being transcribed. + + - `transcript: str` + + The transcribed text. + + - `type: Literal["conversation.item.input_audio_transcription.completed"]` + + The event type, must be + `conversation.item.input_audio_transcription.completed`. + + - `"conversation.item.input_audio_transcription.completed"` + + - `usage: Usage` + + Usage statistics for the transcription, this is billed according to the ASR model's pricing rather than the realtime model's pricing. + + - `class UsageTranscriptTextUsageTokens: …` + + Usage statistics for models billed by token usage. + + - `input_tokens: int` + + Number of input tokens billed for this request. + + - `output_tokens: int` + + Number of output tokens generated. + + - `total_tokens: int` + + Total number of tokens used (input + output). + + - `type: Literal["tokens"]` + + The type of the usage object. Always `tokens` for this variant. + + - `"tokens"` + + - `input_token_details: Optional[UsageTranscriptTextUsageTokensInputTokenDetails]` + + Details about the input tokens billed for this request. + + - `audio_tokens: Optional[int]` + + Number of audio tokens billed for this request. + + - `text_tokens: Optional[int]` + + Number of text tokens billed for this request. + + - `class UsageTranscriptTextUsageDuration: …` + + Usage statistics for models billed by audio input duration. + + - `seconds: float` + + Duration of the input audio in seconds. + + - `type: Literal["duration"]` + + The type of the usage object. Always `duration` for this variant. + + - `"duration"` + + - `logprobs: Optional[List[LogProbProperties]]` + + The log probabilities of the transcription. + + - `token: str` + + The token that was used to generate the log probability. + + - `bytes: List[int]` + + The bytes that were used to generate the log probability. + + - `logprob: float` + + The log probability of the token. + + - `class ConversationItemInputAudioTranscriptionDeltaEvent: …` + + Returned when the text value of an input audio transcription content part is updated with incremental transcription results. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item containing the audio that is being transcribed. + + - `type: Literal["conversation.item.input_audio_transcription.delta"]` + + The event type, must be `conversation.item.input_audio_transcription.delta`. + + - `"conversation.item.input_audio_transcription.delta"` + + - `content_index: Optional[int]` + + The index of the content part in the item's content array. + + - `delta: Optional[str]` + + The text delta. + + - `logprobs: Optional[List[LogProbProperties]]` + + The log probabilities of the transcription. These can be enabled by configurating the session with `"include": ["item.input_audio_transcription.logprobs"]`. Each entry in the array corresponds a log probability of which token would be selected for this chunk of transcription. This can help to identify if it was possible there were multiple valid options for a given chunk of transcription. + + - `token: str` + + The token that was used to generate the log probability. + + - `bytes: List[int]` + + The bytes that were used to generate the log probability. + + - `logprob: float` + + The log probability of the token. + + - `class ConversationItemInputAudioTranscriptionFailedEvent: …` + + Returned when input audio transcription is configured, and a transcription + request for a user message failed. These events are separate from other + `error` events so that the client can identify the related Item. + + - `content_index: int` + + The index of the content part containing the audio. + + - `error: Error` + + Details of the transcription error. + + - `code: Optional[str]` + + Error code, if any. + + - `message: Optional[str]` + + A human-readable error message. + + - `param: Optional[str]` + + Parameter related to the error, if any. + + - `type: Optional[str]` + + The type of error. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the user message item. + + - `type: Literal["conversation.item.input_audio_transcription.failed"]` + + The event type, must be + `conversation.item.input_audio_transcription.failed`. + + - `"conversation.item.input_audio_transcription.failed"` + + - `class ConversationItemRetrieved: …` + + Returned when a conversation item is retrieved with `conversation.item.retrieve`. This is provided as a way to fetch the server's representation of an item, for example to get access to the post-processed audio data after noise cancellation and VAD. It includes the full content of the Item, including audio data. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `type: Literal["conversation.item.retrieved"]` + + The event type, must be `conversation.item.retrieved`. + + - `"conversation.item.retrieved"` + + - `class ConversationItemTruncatedEvent: …` + + Returned when an earlier assistant audio message item is truncated by the + client with a `conversation.item.truncate` event. This event is used to + synchronize the server's understanding of the audio with the client's playback. + + This action will truncate the audio and remove the server-side text transcript + to ensure there is no text in the context that hasn't been heard by the user. + + - `audio_end_ms: int` + + The duration up to which the audio was truncated, in milliseconds. + + - `content_index: int` + + The index of the content part that was truncated. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the assistant message item that was truncated. + + - `type: Literal["conversation.item.truncated"]` + + The event type, must be `conversation.item.truncated`. + + - `"conversation.item.truncated"` + + - `class RealtimeErrorEvent: …` + + Returned when an error occurs, which could be a client problem or a server + problem. Most errors are recoverable and the session will stay open, we + recommend to implementors to monitor and log error messages by default. + + - `error: RealtimeError` + + Details of the error. + + - `message: str` + + A human-readable error message. + + - `type: str` + + The type of error (e.g., "invalid_request_error", "server_error"). + + - `code: Optional[str]` + + Error code, if any. + + - `event_id: Optional[str]` + + The event_id of the client event that caused the error, if applicable. + + - `param: Optional[str]` + + Parameter related to the error, if any. + + - `event_id: str` + + The unique ID of the server event. + + - `type: Literal["error"]` + + The event type, must be `error`. + + - `"error"` + + - `class InputAudioBufferClearedEvent: …` + + Returned when the input audio buffer is cleared by the client with a + `input_audio_buffer.clear` event. + + - `event_id: str` + + The unique ID of the server event. + + - `type: Literal["input_audio_buffer.cleared"]` + + The event type, must be `input_audio_buffer.cleared`. + + - `"input_audio_buffer.cleared"` + + - `class InputAudioBufferCommittedEvent: …` + + Returned when an input audio buffer is committed, either by the client or + automatically in server VAD mode. The `item_id` property is the ID of the user + message item that will be created, thus a `conversation.item.created` event + will also be sent to the client. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the user message item that will be created. + + - `type: Literal["input_audio_buffer.committed"]` + + The event type, must be `input_audio_buffer.committed`. + + - `"input_audio_buffer.committed"` + + - `previous_item_id: Optional[str]` + + The ID of the preceding item after which the new item will be inserted. + Can be `null` if the item has no predecessor. + + - `class InputAudioBufferDtmfEventReceivedEvent: …` + + **SIP Only:** Returned when an DTMF event is received. A DTMF event is a message that + represents a telephone keypad press (0–9, *, #, A–D). The `event` property + is the keypad that the user press. The `received_at` is the UTC Unix Timestamp + that the server received the event. + + - `event: str` + + The telephone keypad that was pressed by the user. + + - `received_at: int` + + UTC Unix Timestamp when DTMF Event was received by server. + + - `type: Literal["input_audio_buffer.dtmf_event_received"]` + + The event type, must be `input_audio_buffer.dtmf_event_received`. + + - `"input_audio_buffer.dtmf_event_received"` + + - `class InputAudioBufferSpeechStartedEvent: …` + + Sent by the server when in `server_vad` mode to indicate that speech has been + detected in the audio buffer. This can happen any time audio is added to the + buffer (unless speech is already detected). The client may want to use this + event to interrupt audio playback or provide visual feedback to the user. + + The client should expect to receive a `input_audio_buffer.speech_stopped` event + when speech stops. The `item_id` property is the ID of the user message item + that will be created when speech stops and will also be included in the + `input_audio_buffer.speech_stopped` event (unless the client manually commits + the audio buffer during VAD activation). + + - `audio_start_ms: int` + + Milliseconds from the start of all audio written to the buffer during the + session when speech was first detected. This will correspond to the + beginning of audio sent to the model, and thus includes the + `prefix_padding_ms` configured in the Session. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the user message item that will be created when speech stops. + + - `type: Literal["input_audio_buffer.speech_started"]` + + The event type, must be `input_audio_buffer.speech_started`. + + - `"input_audio_buffer.speech_started"` + + - `class InputAudioBufferSpeechStoppedEvent: …` + + Returned in `server_vad` mode when the server detects the end of speech in + the audio buffer. The server will also send an `conversation.item.created` + event with the user message item that is created from the audio buffer. + + - `audio_end_ms: int` + + Milliseconds since the session started when speech stopped. This will + correspond to the end of audio sent to the model, and thus includes the + `min_silence_duration_ms` configured in the Session. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the user message item that will be created. + + - `type: Literal["input_audio_buffer.speech_stopped"]` + + The event type, must be `input_audio_buffer.speech_stopped`. + + - `"input_audio_buffer.speech_stopped"` + + - `class RateLimitsUpdatedEvent: …` + + Emitted at the beginning of a Response to indicate the updated rate limits. + When a Response is created some tokens will be "reserved" for the output + tokens, the rate limits shown here reflect that reservation, which is then + adjusted accordingly once the Response is completed. + + - `event_id: str` + + The unique ID of the server event. + + - `rate_limits: List[RateLimit]` + + List of rate limit information. + + - `limit: Optional[int]` + + The maximum allowed value for the rate limit. + + - `name: Optional[Literal["requests", "tokens"]]` + + The name of the rate limit (`requests`, `tokens`). + + - `"requests"` + + - `"tokens"` + + - `remaining: Optional[int]` + + The remaining value before the limit is reached. + + - `reset_seconds: Optional[float]` + + Seconds until the rate limit resets. + + - `type: Literal["rate_limits.updated"]` + + The event type, must be `rate_limits.updated`. + + - `"rate_limits.updated"` + + - `class ResponseAudioDeltaEvent: …` + + Returned when the model-generated audio is updated. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `delta: str` + + Base64-encoded audio data delta. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.output_audio.delta"]` + + The event type, must be `response.output_audio.delta`. + + - `"response.output_audio.delta"` + + - `class ResponseAudioDoneEvent: …` + + Returned when the model-generated audio is done. Also emitted when a Response + is interrupted, incomplete, or cancelled. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.output_audio.done"]` + + The event type, must be `response.output_audio.done`. + + - `"response.output_audio.done"` + + - `class ResponseAudioTranscriptDeltaEvent: …` + + Returned when the model-generated transcription of audio output is updated. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `delta: str` + + The transcript delta. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.output_audio_transcript.delta"]` + + The event type, must be `response.output_audio_transcript.delta`. + + - `"response.output_audio_transcript.delta"` + + - `class ResponseAudioTranscriptDoneEvent: …` + + Returned when the model-generated transcription of audio output is done + streaming. Also emitted when a Response is interrupted, incomplete, or + cancelled. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `transcript: str` + + The final transcript of the audio. + + - `type: Literal["response.output_audio_transcript.done"]` + + The event type, must be `response.output_audio_transcript.done`. + + - `"response.output_audio_transcript.done"` + + - `class ResponseContentPartAddedEvent: …` + + Returned when a new content part is added to an assistant message item during + response generation. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item to which the content part was added. + + - `output_index: int` + + The index of the output item in the response. + + - `part: Part` + + The content part that was added. + + - `audio: Optional[str]` + + Base64-encoded audio data (if type is "audio"). + + - `text: Optional[str]` + + The text content (if type is "text"). + + - `transcript: Optional[str]` + + The transcript of the audio (if type is "audio"). + + - `type: Optional[Literal["text", "audio"]]` + + The content type ("text", "audio"). + + - `"text"` + + - `"audio"` + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.content_part.added"]` + + The event type, must be `response.content_part.added`. + + - `"response.content_part.added"` + + - `class ResponseContentPartDoneEvent: …` + + Returned when a content part is done streaming in an assistant message item. + Also emitted when a Response is interrupted, incomplete, or cancelled. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `part: Part` + + The content part that is done. + + - `audio: Optional[str]` + + Base64-encoded audio data (if type is "audio"). + + - `text: Optional[str]` + + The text content (if type is "text"). + + - `transcript: Optional[str]` + + The transcript of the audio (if type is "audio"). + + - `type: Optional[Literal["text", "audio"]]` + + The content type ("text", "audio"). + + - `"text"` + + - `"audio"` + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.content_part.done"]` + + The event type, must be `response.content_part.done`. + + - `"response.content_part.done"` + + - `class ResponseCreatedEvent: …` + + Returned when a new Response is created. The first event of response creation, + where the response is in an initial state of `in_progress`. + + - `event_id: str` + + The unique ID of the server event. + + - `response: RealtimeResponse` + + The response resource. + + - `id: Optional[str]` + + The unique ID of the response, will look like `resp_1234`. + + - `audio: Optional[Audio]` + + Configuration for audio output. + + - `output: Optional[AudioOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `voice: Optional[Union[str, Literal["alloy", "ash", "ballad", 7 more], null]]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `conversation_id: Optional[str]` + + Which conversation the response is added to, determined by the `conversation` + field in the `response.create` event. If `auto`, the response will be added to + the default conversation and the value of `conversation_id` will be an id like + `conv_1234`. If `none`, the response will not be added to any conversation and + the value of `conversation_id` will be `null`. If responses are being triggered + automatically by VAD the response will be added to the default conversation + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls, that was used in this response. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `metadata: Optional[Metadata]` + + Set of 16 key-value pairs that can be attached to an object. This can be + useful for storing additional information about the object in a structured + format, and querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings + with a maximum length of 512 characters. + + - `object: Optional[Literal["realtime.response"]]` + + The object type, must be `realtime.response`. + + - `"realtime.response"` + + - `output: Optional[List[ConversationItem]]` + + The list of output items generated by the response. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model used to respond, currently the only possible values are + `[\"audio\"]`, `[\"text\"]`. Audio output always include a text transcript. Setting the + output to mode `text` will disable audio output from the model. + + - `"text"` + + - `"audio"` + + - `status: Optional[Literal["completed", "cancelled", "failed", 2 more]]` + + The final status of the response (`completed`, `cancelled`, `failed`, or + `incomplete`, `in_progress`). + + - `"completed"` + + - `"cancelled"` + + - `"failed"` + + - `"incomplete"` + + - `"in_progress"` + + - `status_details: Optional[RealtimeResponseStatus]` + + Additional details about the status. + + - `error: Optional[Error]` + + A description of the error that caused the response to fail, + populated when the `status` is `failed`. + + - `code: Optional[str]` + + Error code, if any. + + - `type: Optional[str]` + + The type of error. + + - `reason: Optional[Literal["turn_detected", "client_cancelled", "max_output_tokens", "content_filter"]]` + + The reason the Response did not complete. For a `cancelled` Response, one of `turn_detected` (the server VAD detected a new start of speech) or `client_cancelled` (the client sent a cancel event). For an `incomplete` Response, one of `max_output_tokens` or `content_filter` (the server-side safety filter activated and cut off the response). + + - `"turn_detected"` + + - `"client_cancelled"` + + - `"max_output_tokens"` + + - `"content_filter"` + + - `type: Optional[Literal["completed", "cancelled", "incomplete", "failed"]]` + + The type of error that caused the response to fail, corresponding + with the `status` field (`completed`, `cancelled`, `incomplete`, + `failed`). + + - `"completed"` + + - `"cancelled"` + + - `"incomplete"` + + - `"failed"` + + - `usage: Optional[RealtimeResponseUsage]` + + Usage statistics for the Response, this will correspond to billing. A + Realtime API session will maintain a conversation context and append new + Items to the Conversation, thus output from previous turns (text and + audio tokens) will become the input for later turns. + + - `input_token_details: Optional[RealtimeResponseUsageInputTokenDetails]` + + Details about the input tokens used in the Response. Cached tokens are tokens from previous turns in the conversation that are included as context for the current response. Cached tokens here are counted as a subset of input tokens, meaning input tokens will include cached and uncached tokens. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used as input for the Response. + + - `cached_tokens: Optional[int]` + + The number of cached tokens used as input for the Response. + + - `cached_tokens_details: Optional[CachedTokensDetails]` + + Details about the cached tokens used as input for the Response. + + - `audio_tokens: Optional[int]` + + The number of cached audio tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of cached image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of cached text tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used as input for the Response. + + - `input_tokens: Optional[int]` + + The number of input tokens used in the Response, including text and + audio tokens. + + - `output_token_details: Optional[RealtimeResponseUsageOutputTokenDetails]` + + Details about the output tokens used in the Response. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used in the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used in the Response. + + - `output_tokens: Optional[int]` + + The number of output tokens sent in the Response, including text and + audio tokens. + + - `total_tokens: Optional[int]` + + The total number of tokens in the Response including input and output + text and audio tokens. + + - `type: Literal["response.created"]` + + The event type, must be `response.created`. + + - `"response.created"` + + - `class ResponseDoneEvent: …` + + Returned when a Response is done streaming. Always emitted, no matter the + final state. The Response object included in the `response.done` event will + include all output Items in the Response but will omit the raw audio data. + + Clients should check the `status` field of the Response to determine if it was successful + (`completed`) or if there was another outcome: `cancelled`, `failed`, or `incomplete`. + + A response will contain all output items that were generated during the response, excluding + any audio content. + + - `event_id: str` + + The unique ID of the server event. + + - `response: RealtimeResponse` + + The response resource. + + - `type: Literal["response.done"]` + + The event type, must be `response.done`. + + - `"response.done"` + + - `class ResponseFunctionCallArgumentsDeltaEvent: …` + + Returned when the model-generated function call arguments are updated. + + - `call_id: str` + + The ID of the function call. + + - `delta: str` + + The arguments delta as a JSON string. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the function call item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.function_call_arguments.delta"]` + + The event type, must be `response.function_call_arguments.delta`. + + - `"response.function_call_arguments.delta"` + + - `class ResponseFunctionCallArgumentsDoneEvent: …` + + Returned when the model-generated function call arguments are done streaming. + Also emitted when a Response is interrupted, incomplete, or cancelled. + + - `arguments: str` + + The final arguments as a JSON string. + + - `call_id: str` + + The ID of the function call. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the function call item. + + - `name: str` + + The name of the function that was called. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.function_call_arguments.done"]` + + The event type, must be `response.function_call_arguments.done`. + + - `"response.function_call_arguments.done"` + + - `class ResponseOutputItemAddedEvent: …` + + Returned when a new Item is created during Response generation. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `output_index: int` + + The index of the output item in the Response. + + - `response_id: str` + + The ID of the Response to which the item belongs. + + - `type: Literal["response.output_item.added"]` + + The event type, must be `response.output_item.added`. + + - `"response.output_item.added"` + + - `class ResponseOutputItemDoneEvent: …` + + Returned when an Item is done streaming. Also emitted when a Response is + interrupted, incomplete, or cancelled. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `output_index: int` + + The index of the output item in the Response. + + - `response_id: str` + + The ID of the Response to which the item belongs. + + - `type: Literal["response.output_item.done"]` + + The event type, must be `response.output_item.done`. + + - `"response.output_item.done"` + + - `class ResponseTextDeltaEvent: …` + + Returned when the text value of an "output_text" content part is updated. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `delta: str` + + The text delta. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.output_text.delta"]` + + The event type, must be `response.output_text.delta`. + + - `"response.output_text.delta"` + + - `class ResponseTextDoneEvent: …` + + Returned when the text value of an "output_text" content part is done streaming. Also + emitted when a Response is interrupted, incomplete, or cancelled. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `text: str` + + The final text content. + + - `type: Literal["response.output_text.done"]` + + The event type, must be `response.output_text.done`. + + - `"response.output_text.done"` + + - `class SessionCreatedEvent: …` + + Returned when a Session is created. Emitted automatically when a new + connection is established as the first server event. This event will contain + the default Session configuration. + + - `event_id: str` + + The unique ID of the server event. + + - `session: Session` + + The session configuration. + + - `class RealtimeSessionCreateRequest: …` + + Realtime session object configuration. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[RealtimeAudioConfig]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[RealtimeToolChoiceConfig]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[RealtimeToolsConfig]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + + - `class RealtimeTranscriptionSessionCreateRequest: …` + + Realtime transcription session object configuration. + + - `type: Literal["transcription"]` + + The type of session to create. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[RealtimeTranscriptionSessionAudio]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeTranscriptionSessionAudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `type: Literal["session.created"]` + + The event type, must be `session.created`. + + - `"session.created"` + + - `class SessionUpdatedEvent: …` + + Returned when a session is updated with a `session.update` event, unless + there is an error. + + - `event_id: str` + + The unique ID of the server event. + + - `session: Session` + + The session configuration. + + - `class RealtimeSessionCreateRequest: …` + + Realtime session object configuration. + + - `class RealtimeTranscriptionSessionCreateRequest: …` + + Realtime transcription session object configuration. + + - `type: Literal["session.updated"]` + + The event type, must be `session.updated`. + + - `"session.updated"` + + - `class OutputAudioBufferStarted: …` + + **WebRTC/SIP Only:** Emitted when the server begins streaming audio to the client. This event is + emitted after an audio content part has been added (`response.content_part.added`) + to the response. + [Learn more](https://platform.openai.com/docs/guides/realtime-conversations#client-and-server-events-for-audio-in-webrtc). + + - `event_id: str` + + The unique ID of the server event. + + - `response_id: str` + + The unique ID of the response that produced the audio. + + - `type: Literal["output_audio_buffer.started"]` + + The event type, must be `output_audio_buffer.started`. + + - `"output_audio_buffer.started"` + + - `class OutputAudioBufferStopped: …` + + **WebRTC/SIP Only:** Emitted when the output audio buffer has been completely drained on the server, + and no more audio is forthcoming. This event is emitted after the full response + data has been sent to the client (`response.done`). + [Learn more](https://platform.openai.com/docs/guides/realtime-conversations#client-and-server-events-for-audio-in-webrtc). + + - `event_id: str` + + The unique ID of the server event. + + - `response_id: str` + + The unique ID of the response that produced the audio. + + - `type: Literal["output_audio_buffer.stopped"]` + + The event type, must be `output_audio_buffer.stopped`. + + - `"output_audio_buffer.stopped"` + + - `class OutputAudioBufferCleared: …` + + **WebRTC/SIP Only:** Emitted when the output audio buffer is cleared. This happens either in VAD + mode when the user has interrupted (`input_audio_buffer.speech_started`), + or when the client has emitted the `output_audio_buffer.clear` event to manually + cut off the current audio response. + [Learn more](https://platform.openai.com/docs/guides/realtime-conversations#client-and-server-events-for-audio-in-webrtc). + + - `event_id: str` + + The unique ID of the server event. + + - `response_id: str` + + The unique ID of the response that produced the audio. + + - `type: Literal["output_audio_buffer.cleared"]` + + The event type, must be `output_audio_buffer.cleared`. + + - `"output_audio_buffer.cleared"` + + - `class ConversationItemAdded: …` + + Sent by the server when an Item is added to the default Conversation. This can happen in several cases: + + - When the client sends a `conversation.item.create` event. + - When the input audio buffer is committed. In this case the item will be a user message containing the audio from the buffer. + - When the model is generating a Response. In this case the `conversation.item.added` event will be sent when the model starts generating a specific Item, and thus it will not yet have any content (and `status` will be `in_progress`). + + The event will include the full content of the Item (except when model is generating a Response) except for audio data, which can be retrieved separately with a `conversation.item.retrieve` event if necessary. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `type: Literal["conversation.item.added"]` + + The event type, must be `conversation.item.added`. + + - `"conversation.item.added"` + + - `previous_item_id: Optional[str]` + + The ID of the item that precedes this one, if any. This is used to + maintain ordering when items are inserted. + + - `class ConversationItemDone: …` + + Returned when a conversation item is finalized. + + The event will include the full content of the Item except for audio data, which can be retrieved separately with a `conversation.item.retrieve` event if needed. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `type: Literal["conversation.item.done"]` + + The event type, must be `conversation.item.done`. + + - `"conversation.item.done"` + + - `previous_item_id: Optional[str]` + + The ID of the item that precedes this one, if any. This is used to + maintain ordering when items are inserted. + + - `class InputAudioBufferTimeoutTriggered: …` + + Returned when the Server VAD timeout is triggered for the input audio buffer. This is configured + with `idle_timeout_ms` in the `turn_detection` settings of the session, and it indicates that + there hasn't been any speech detected for the configured duration. + + The `audio_start_ms` and `audio_end_ms` fields indicate the segment of audio after the last + model response up to the triggering time, as an offset from the beginning of audio written + to the input audio buffer. This means it demarcates the segment of audio that was silent and + the difference between the start and end values will roughly match the configured timeout. + + The empty audio will be committed to the conversation as an `input_audio` item (there will be a + `input_audio_buffer.committed` event) and a model response will be generated. There may be speech + that didn't trigger VAD but is still detected by the model, so the model may respond with + something relevant to the conversation or a prompt to continue speaking. + + - `audio_end_ms: int` + + Millisecond offset of audio written to the input audio buffer at the time the timeout was triggered. + + - `audio_start_ms: int` + + Millisecond offset of audio written to the input audio buffer that was after the playback time of the last model response. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item associated with this segment. + + - `type: Literal["input_audio_buffer.timeout_triggered"]` + + The event type, must be `input_audio_buffer.timeout_triggered`. + + - `"input_audio_buffer.timeout_triggered"` + + - `class ConversationItemInputAudioTranscriptionSegment: …` + + Returned when an input audio transcription segment is identified for an item. + + - `id: str` + + The segment identifier. + + - `content_index: int` + + The index of the input audio content part within the item. + + - `end: float` + + End time of the segment in seconds. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item containing the input audio content. + + - `speaker: str` + + The detected speaker label for this segment. + + - `start: float` + + Start time of the segment in seconds. + + - `text: str` + + The text for this segment. + + - `type: Literal["conversation.item.input_audio_transcription.segment"]` + + The event type, must be `conversation.item.input_audio_transcription.segment`. + + - `"conversation.item.input_audio_transcription.segment"` + + - `class McpListToolsInProgress: …` + + Returned when listing MCP tools is in progress for an item. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP list tools item. + + - `type: Literal["mcp_list_tools.in_progress"]` + + The event type, must be `mcp_list_tools.in_progress`. + + - `"mcp_list_tools.in_progress"` + + - `class McpListToolsCompleted: …` + + Returned when listing MCP tools has completed for an item. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP list tools item. + + - `type: Literal["mcp_list_tools.completed"]` + + The event type, must be `mcp_list_tools.completed`. + + - `"mcp_list_tools.completed"` + + - `class McpListToolsFailed: …` + + Returned when listing MCP tools has failed for an item. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP list tools item. + + - `type: Literal["mcp_list_tools.failed"]` + + The event type, must be `mcp_list_tools.failed`. + + - `"mcp_list_tools.failed"` + + - `class ResponseMcpCallArgumentsDelta: …` + + Returned when MCP tool call arguments are updated during response generation. + + - `delta: str` + + The JSON-encoded arguments delta. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.mcp_call_arguments.delta"]` + + The event type, must be `response.mcp_call_arguments.delta`. + + - `"response.mcp_call_arguments.delta"` + + - `obfuscation: Optional[str]` + + If present, indicates the delta text was obfuscated. + + - `class ResponseMcpCallArgumentsDone: …` + + Returned when MCP tool call arguments are finalized during response generation. + + - `arguments: str` + + The final JSON-encoded arguments string. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.mcp_call_arguments.done"]` + + The event type, must be `response.mcp_call_arguments.done`. + + - `"response.mcp_call_arguments.done"` + + - `class ResponseMcpCallInProgress: …` + + Returned when an MCP tool call has started and is in progress. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `type: Literal["response.mcp_call.in_progress"]` + + The event type, must be `response.mcp_call.in_progress`. + + - `"response.mcp_call.in_progress"` + + - `class ResponseMcpCallCompleted: …` + + Returned when an MCP tool call has completed successfully. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `type: Literal["response.mcp_call.completed"]` + + The event type, must be `response.mcp_call.completed`. + + - `"response.mcp_call.completed"` + + - `class ResponseMcpCallFailed: …` + + Returned when an MCP tool call has failed. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `type: Literal["response.mcp_call.failed"]` + + The event type, must be `response.mcp_call.failed`. + + - `"response.mcp_call.failed"` + +### Realtime Session + +- `class RealtimeSession: …` + + Realtime session object for the beta interface. + + - `id: Optional[str]` + + Unique identifier for the session that looks like `sess_1234567890abcdef`. + + - `expires_at: Optional[int]` + + Expiration timestamp for the session, in seconds since epoch. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + - `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `input_audio_format: Optional[Literal["pcm16", "g711_ulaw", "g711_alaw"]]` + + The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. + For `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate, + single channel (mono), and little-endian byte order. + + - `"pcm16"` + + - `"g711_ulaw"` + + - `"g711_alaw"` + + - `input_audio_noise_reduction: Optional[InputAudioNoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `input_audio_transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model + calls. This field allows the client to guide the model on desired + responses. The model can be instructed on response content and format, + (e.g. "be extremely succinct", "act friendly", "here are examples of good + responses") and on audio behavior (e.g. "talk quickly", "inject emotion + into your voice", "laugh frequently"). The instructions are not + guaranteed to be followed by the model, but they provide guidance to the + model on the desired behavior. + + Note that the server sets default instructions which will be used if this + field is not set and are visible in the `session.created` event at the + start of the session. + + - `max_response_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. To disable audio, + set this to ["text"]. + + - `"text"` + + - `"audio"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `object: Optional[Literal["realtime.session"]]` + + The object type. Always `realtime.session`. + + - `"realtime.session"` + + - `output_audio_format: Optional[Literal["pcm16", "g711_ulaw", "g711_alaw"]]` + + The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. + For `pcm16`, output audio is sampled at a rate of 24kHz. + + - `"pcm16"` + + - `"g711_ulaw"` + + - `"g711_alaw"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `speed: Optional[float]` + + The speed of the model's spoken response. 1.0 is the default speed. 0.25 is + the minimum speed. 1.5 is the maximum speed. This value can only be changed + in between model turns, not while a response is in progress. + + - `temperature: Optional[float]` + + Sampling temperature for the model, limited to [0.6, 1.2]. For audio models a temperature of 0.8 is highly recommended for best performance. + + - `tool_choice: Optional[str]` + + How the model chooses tools. Options are `auto`, `none`, `required`, or + specify a function. + + - `tools: Optional[List[RealtimeFunctionTool]]` + + Tools (functions) available to the model. + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `tracing: Optional[Tracing]` + + Configuration options for tracing. Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Default tracing mode for the session. + + - `"auto"` + + - `class TracingTracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the traces dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the traces dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the traces dashboard. + + - `turn_detection: Optional[TurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class TurnDetectionServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class TurnDetectionSemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `voice: Optional[Union[str, Literal["alloy", "ash", "ballad", 7 more], null]]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, and `verse`. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, and `verse`. + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + +### Realtime Session Create Request + +- `class RealtimeSessionCreateRequest: …` + + Realtime session object configuration. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[RealtimeAudioConfig]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[RealtimeToolChoiceConfig]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[RealtimeToolsConfig]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + +### Realtime Tool Choice Config + +- `RealtimeToolChoiceConfig` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + +### Realtime Tools Config + +- `List[RealtimeToolsConfigUnion]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + +### Realtime Tools Config Union + +- `RealtimeToolsConfigUnion` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + +### Realtime Tracing Config + +- `Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + +### Realtime Transcription Session Audio + +- `class RealtimeTranscriptionSessionAudio: …` + + Configuration for input and output audio. + + - `input: Optional[RealtimeTranscriptionSessionAudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + +### Realtime Transcription Session Audio Input + +- `class RealtimeTranscriptionSessionAudioInput: …` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + +### Realtime Transcription Session Audio Input Turn Detection + +- `Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + +### Realtime Transcription Session Create Request + +- `class RealtimeTranscriptionSessionCreateRequest: …` + + Realtime transcription session object configuration. + + - `type: Literal["transcription"]` + + The type of session to create. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[RealtimeTranscriptionSessionAudio]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeTranscriptionSessionAudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + +### Realtime Truncation + +- `RealtimeTruncation` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + +### Realtime Truncation Retention Ratio + +- `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + +### Response Audio Delta Event + +- `class ResponseAudioDeltaEvent: …` + + Returned when the model-generated audio is updated. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `delta: str` + + Base64-encoded audio data delta. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.output_audio.delta"]` + + The event type, must be `response.output_audio.delta`. + + - `"response.output_audio.delta"` + +### Response Audio Done Event + +- `class ResponseAudioDoneEvent: …` + + Returned when the model-generated audio is done. Also emitted when a Response + is interrupted, incomplete, or cancelled. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.output_audio.done"]` + + The event type, must be `response.output_audio.done`. + + - `"response.output_audio.done"` + +### Response Audio Transcript Delta Event + +- `class ResponseAudioTranscriptDeltaEvent: …` + + Returned when the model-generated transcription of audio output is updated. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `delta: str` + + The transcript delta. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.output_audio_transcript.delta"]` + + The event type, must be `response.output_audio_transcript.delta`. + + - `"response.output_audio_transcript.delta"` + +### Response Audio Transcript Done Event + +- `class ResponseAudioTranscriptDoneEvent: …` + + Returned when the model-generated transcription of audio output is done + streaming. Also emitted when a Response is interrupted, incomplete, or + cancelled. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `transcript: str` + + The final transcript of the audio. + + - `type: Literal["response.output_audio_transcript.done"]` + + The event type, must be `response.output_audio_transcript.done`. + + - `"response.output_audio_transcript.done"` + +### Response Cancel Event + +- `class ResponseCancelEvent: …` + + Send this event to cancel an in-progress response. The server will respond + with a `response.done` event with a status of `response.status=cancelled`. If + there is no response to cancel, the server will respond with an error. It's safe + to call `response.cancel` even if no response is in progress, an error will be + returned the session will remain unaffected. + + - `type: Literal["response.cancel"]` + + The event type, must be `response.cancel`. + + - `"response.cancel"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `response_id: Optional[str]` + + A specific response ID to cancel - if not provided, will cancel an + in-progress response in the default conversation. + +### Response Content Part Added Event + +- `class ResponseContentPartAddedEvent: …` + + Returned when a new content part is added to an assistant message item during + response generation. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item to which the content part was added. + + - `output_index: int` + + The index of the output item in the response. + + - `part: Part` + + The content part that was added. + + - `audio: Optional[str]` + + Base64-encoded audio data (if type is "audio"). + + - `text: Optional[str]` + + The text content (if type is "text"). + + - `transcript: Optional[str]` + + The transcript of the audio (if type is "audio"). + + - `type: Optional[Literal["text", "audio"]]` + + The content type ("text", "audio"). + + - `"text"` + + - `"audio"` + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.content_part.added"]` + + The event type, must be `response.content_part.added`. + + - `"response.content_part.added"` + +### Response Content Part Done Event + +- `class ResponseContentPartDoneEvent: …` + + Returned when a content part is done streaming in an assistant message item. + Also emitted when a Response is interrupted, incomplete, or cancelled. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `part: Part` + + The content part that is done. + + - `audio: Optional[str]` + + Base64-encoded audio data (if type is "audio"). + + - `text: Optional[str]` + + The text content (if type is "text"). + + - `transcript: Optional[str]` + + The transcript of the audio (if type is "audio"). + + - `type: Optional[Literal["text", "audio"]]` + + The content type ("text", "audio"). + + - `"text"` + + - `"audio"` + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.content_part.done"]` + + The event type, must be `response.content_part.done`. + + - `"response.content_part.done"` + +### Response Create Event + +- `class ResponseCreateEvent: …` + + This event instructs the server to create a Response, which means triggering + model inference. When in Server VAD mode, the server will create Responses + automatically. + + A Response will include at least one Item, and may have two, in which case + the second will be a function call. These Items will be appended to the + conversation history by default. + + The server will respond with a `response.created` event, events for Items + and content created, and finally a `response.done` event to indicate the + Response is complete. + + The `response.create` event includes inference configuration like + `instructions` and `tools`. If these are set, they will override the Session's + configuration for this Response only. + + Responses can be created out-of-band of the default Conversation, meaning that they can + have arbitrary input, and it's possible to disable writing the output to the Conversation. + Only one Response can write to the default Conversation at a time, but otherwise multiple + Responses can be created in parallel. The `metadata` field is a good way to disambiguate + multiple simultaneous Responses. + + Clients can set `conversation` to `none` to create a Response that does not write to the default + Conversation. Arbitrary input can be provided with the `input` field, which is an array accepting + raw Items and references to existing Items. + + - `type: Literal["response.create"]` + + The event type, must be `response.create`. + + - `"response.create"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + + - `response: Optional[RealtimeResponseCreateParams]` + + Create a new Realtime response with these parameters + + - `audio: Optional[RealtimeResponseCreateAudioOutput]` + + Configuration for audio input and output. + + - `output: Optional[Output]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `voice: Optional[OutputVoice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class OutputVoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `conversation: Optional[Union[str, Literal["auto", "none"], null]]` + + Controls which conversation the response is added to. Currently supports + `auto` and `none`, with `auto` as the default value. The `auto` value + means that the contents of the response will be added to the default + conversation. Set this to `none` to create an out-of-band response which + will not add items to default conversation. + + - `str` + + - `Literal["auto", "none"]` + + Controls which conversation the response is added to. Currently supports + `auto` and `none`, with `auto` as the default value. The `auto` value + means that the contents of the response will be added to the default + conversation. Set this to `none` to create an out-of-band response which + will not add items to default conversation. + + - `"auto"` + + - `"none"` + + - `input: Optional[List[ConversationItem]]` + + Input items to include in the prompt for the model. Using this field + creates a new context for this Response instead of using the default + conversation. An empty array `[]` will clear the context for this Response. + Note that this can include references to items that previously appeared in the session + using their id. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `metadata: Optional[Metadata]` + + Set of 16 key-value pairs that can be attached to an object. This can be + useful for storing additional information about the object in a structured + format, and querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings + with a maximum length of 512 characters. + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model used to respond, currently the only possible values are + `[\"audio\"]`, `[\"text\"]`. Audio output always include a text transcript. Setting the + output to mode `text` will disable audio output from the model. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[ToolChoice]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[List[Tool]]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class RealtimeResponseCreateMcpTool: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[AllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class AllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[RequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class RequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[RequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[RequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + +### Response Created Event + +- `class ResponseCreatedEvent: …` + + Returned when a new Response is created. The first event of response creation, + where the response is in an initial state of `in_progress`. + + - `event_id: str` + + The unique ID of the server event. + + - `response: RealtimeResponse` + + The response resource. + + - `id: Optional[str]` + + The unique ID of the response, will look like `resp_1234`. + + - `audio: Optional[Audio]` + + Configuration for audio output. + + - `output: Optional[AudioOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `voice: Optional[Union[str, Literal["alloy", "ash", "ballad", 7 more], null]]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `conversation_id: Optional[str]` + + Which conversation the response is added to, determined by the `conversation` + field in the `response.create` event. If `auto`, the response will be added to + the default conversation and the value of `conversation_id` will be an id like + `conv_1234`. If `none`, the response will not be added to any conversation and + the value of `conversation_id` will be `null`. If responses are being triggered + automatically by VAD the response will be added to the default conversation + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls, that was used in this response. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `metadata: Optional[Metadata]` + + Set of 16 key-value pairs that can be attached to an object. This can be + useful for storing additional information about the object in a structured + format, and querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings + with a maximum length of 512 characters. + + - `object: Optional[Literal["realtime.response"]]` + + The object type, must be `realtime.response`. + + - `"realtime.response"` + + - `output: Optional[List[ConversationItem]]` + + The list of output items generated by the response. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model used to respond, currently the only possible values are + `[\"audio\"]`, `[\"text\"]`. Audio output always include a text transcript. Setting the + output to mode `text` will disable audio output from the model. + + - `"text"` + + - `"audio"` + + - `status: Optional[Literal["completed", "cancelled", "failed", 2 more]]` + + The final status of the response (`completed`, `cancelled`, `failed`, or + `incomplete`, `in_progress`). + + - `"completed"` + + - `"cancelled"` + + - `"failed"` + + - `"incomplete"` + + - `"in_progress"` + + - `status_details: Optional[RealtimeResponseStatus]` + + Additional details about the status. + + - `error: Optional[Error]` + + A description of the error that caused the response to fail, + populated when the `status` is `failed`. + + - `code: Optional[str]` + + Error code, if any. + + - `type: Optional[str]` + + The type of error. + + - `reason: Optional[Literal["turn_detected", "client_cancelled", "max_output_tokens", "content_filter"]]` + + The reason the Response did not complete. For a `cancelled` Response, one of `turn_detected` (the server VAD detected a new start of speech) or `client_cancelled` (the client sent a cancel event). For an `incomplete` Response, one of `max_output_tokens` or `content_filter` (the server-side safety filter activated and cut off the response). + + - `"turn_detected"` + + - `"client_cancelled"` + + - `"max_output_tokens"` + + - `"content_filter"` + + - `type: Optional[Literal["completed", "cancelled", "incomplete", "failed"]]` + + The type of error that caused the response to fail, corresponding + with the `status` field (`completed`, `cancelled`, `incomplete`, + `failed`). + + - `"completed"` + + - `"cancelled"` + + - `"incomplete"` + + - `"failed"` + + - `usage: Optional[RealtimeResponseUsage]` + + Usage statistics for the Response, this will correspond to billing. A + Realtime API session will maintain a conversation context and append new + Items to the Conversation, thus output from previous turns (text and + audio tokens) will become the input for later turns. + + - `input_token_details: Optional[RealtimeResponseUsageInputTokenDetails]` + + Details about the input tokens used in the Response. Cached tokens are tokens from previous turns in the conversation that are included as context for the current response. Cached tokens here are counted as a subset of input tokens, meaning input tokens will include cached and uncached tokens. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used as input for the Response. + + - `cached_tokens: Optional[int]` + + The number of cached tokens used as input for the Response. + + - `cached_tokens_details: Optional[CachedTokensDetails]` + + Details about the cached tokens used as input for the Response. + + - `audio_tokens: Optional[int]` + + The number of cached audio tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of cached image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of cached text tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used as input for the Response. + + - `input_tokens: Optional[int]` + + The number of input tokens used in the Response, including text and + audio tokens. + + - `output_token_details: Optional[RealtimeResponseUsageOutputTokenDetails]` + + Details about the output tokens used in the Response. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used in the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used in the Response. + + - `output_tokens: Optional[int]` + + The number of output tokens sent in the Response, including text and + audio tokens. + + - `total_tokens: Optional[int]` + + The total number of tokens in the Response including input and output + text and audio tokens. + + - `type: Literal["response.created"]` + + The event type, must be `response.created`. + + - `"response.created"` + +### Response Done Event + +- `class ResponseDoneEvent: …` + + Returned when a Response is done streaming. Always emitted, no matter the + final state. The Response object included in the `response.done` event will + include all output Items in the Response but will omit the raw audio data. + + Clients should check the `status` field of the Response to determine if it was successful + (`completed`) or if there was another outcome: `cancelled`, `failed`, or `incomplete`. + + A response will contain all output items that were generated during the response, excluding + any audio content. + + - `event_id: str` + + The unique ID of the server event. + + - `response: RealtimeResponse` + + The response resource. + + - `id: Optional[str]` + + The unique ID of the response, will look like `resp_1234`. + + - `audio: Optional[Audio]` + + Configuration for audio output. + + - `output: Optional[AudioOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `voice: Optional[Union[str, Literal["alloy", "ash", "ballad", 7 more], null]]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `conversation_id: Optional[str]` + + Which conversation the response is added to, determined by the `conversation` + field in the `response.create` event. If `auto`, the response will be added to + the default conversation and the value of `conversation_id` will be an id like + `conv_1234`. If `none`, the response will not be added to any conversation and + the value of `conversation_id` will be `null`. If responses are being triggered + automatically by VAD the response will be added to the default conversation + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls, that was used in this response. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `metadata: Optional[Metadata]` + + Set of 16 key-value pairs that can be attached to an object. This can be + useful for storing additional information about the object in a structured + format, and querying for objects via API or the dashboard. + + Keys are strings with a maximum length of 64 characters. Values are strings + with a maximum length of 512 characters. + + - `object: Optional[Literal["realtime.response"]]` + + The object type, must be `realtime.response`. + + - `"realtime.response"` + + - `output: Optional[List[ConversationItem]]` + + The list of output items generated by the response. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model used to respond, currently the only possible values are + `[\"audio\"]`, `[\"text\"]`. Audio output always include a text transcript. Setting the + output to mode `text` will disable audio output from the model. + + - `"text"` + + - `"audio"` + + - `status: Optional[Literal["completed", "cancelled", "failed", 2 more]]` + + The final status of the response (`completed`, `cancelled`, `failed`, or + `incomplete`, `in_progress`). + + - `"completed"` + + - `"cancelled"` + + - `"failed"` + + - `"incomplete"` + + - `"in_progress"` + + - `status_details: Optional[RealtimeResponseStatus]` + + Additional details about the status. + + - `error: Optional[Error]` + + A description of the error that caused the response to fail, + populated when the `status` is `failed`. + + - `code: Optional[str]` + + Error code, if any. + + - `type: Optional[str]` + + The type of error. + + - `reason: Optional[Literal["turn_detected", "client_cancelled", "max_output_tokens", "content_filter"]]` + + The reason the Response did not complete. For a `cancelled` Response, one of `turn_detected` (the server VAD detected a new start of speech) or `client_cancelled` (the client sent a cancel event). For an `incomplete` Response, one of `max_output_tokens` or `content_filter` (the server-side safety filter activated and cut off the response). + + - `"turn_detected"` + + - `"client_cancelled"` + + - `"max_output_tokens"` + + - `"content_filter"` + + - `type: Optional[Literal["completed", "cancelled", "incomplete", "failed"]]` + + The type of error that caused the response to fail, corresponding + with the `status` field (`completed`, `cancelled`, `incomplete`, + `failed`). + + - `"completed"` + + - `"cancelled"` + + - `"incomplete"` + + - `"failed"` + + - `usage: Optional[RealtimeResponseUsage]` + + Usage statistics for the Response, this will correspond to billing. A + Realtime API session will maintain a conversation context and append new + Items to the Conversation, thus output from previous turns (text and + audio tokens) will become the input for later turns. + + - `input_token_details: Optional[RealtimeResponseUsageInputTokenDetails]` + + Details about the input tokens used in the Response. Cached tokens are tokens from previous turns in the conversation that are included as context for the current response. Cached tokens here are counted as a subset of input tokens, meaning input tokens will include cached and uncached tokens. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used as input for the Response. + + - `cached_tokens: Optional[int]` + + The number of cached tokens used as input for the Response. + + - `cached_tokens_details: Optional[CachedTokensDetails]` + + Details about the cached tokens used as input for the Response. + + - `audio_tokens: Optional[int]` + + The number of cached audio tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of cached image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of cached text tokens used as input for the Response. + + - `image_tokens: Optional[int]` + + The number of image tokens used as input for the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used as input for the Response. + + - `input_tokens: Optional[int]` + + The number of input tokens used in the Response, including text and + audio tokens. + + - `output_token_details: Optional[RealtimeResponseUsageOutputTokenDetails]` + + Details about the output tokens used in the Response. + + - `audio_tokens: Optional[int]` + + The number of audio tokens used in the Response. + + - `text_tokens: Optional[int]` + + The number of text tokens used in the Response. + + - `output_tokens: Optional[int]` + + The number of output tokens sent in the Response, including text and + audio tokens. + + - `total_tokens: Optional[int]` + + The total number of tokens in the Response including input and output + text and audio tokens. + + - `type: Literal["response.done"]` + + The event type, must be `response.done`. + + - `"response.done"` + +### Response Function Call Arguments Delta Event + +- `class ResponseFunctionCallArgumentsDeltaEvent: …` + + Returned when the model-generated function call arguments are updated. + + - `call_id: str` + + The ID of the function call. + + - `delta: str` + + The arguments delta as a JSON string. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the function call item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.function_call_arguments.delta"]` + + The event type, must be `response.function_call_arguments.delta`. + + - `"response.function_call_arguments.delta"` + +### Response Function Call Arguments Done Event + +- `class ResponseFunctionCallArgumentsDoneEvent: …` + + Returned when the model-generated function call arguments are done streaming. + Also emitted when a Response is interrupted, incomplete, or cancelled. + + - `arguments: str` + + The final arguments as a JSON string. + + - `call_id: str` + + The ID of the function call. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the function call item. + + - `name: str` + + The name of the function that was called. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.function_call_arguments.done"]` + + The event type, must be `response.function_call_arguments.done`. + + - `"response.function_call_arguments.done"` + +### Response Mcp Call Arguments Delta + +- `class ResponseMcpCallArgumentsDelta: …` + + Returned when MCP tool call arguments are updated during response generation. + + - `delta: str` + + The JSON-encoded arguments delta. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.mcp_call_arguments.delta"]` + + The event type, must be `response.mcp_call_arguments.delta`. + + - `"response.mcp_call_arguments.delta"` + + - `obfuscation: Optional[str]` + + If present, indicates the delta text was obfuscated. + +### Response Mcp Call Arguments Done + +- `class ResponseMcpCallArgumentsDone: …` + + Returned when MCP tool call arguments are finalized during response generation. + + - `arguments: str` + + The final JSON-encoded arguments string. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.mcp_call_arguments.done"]` + + The event type, must be `response.mcp_call_arguments.done`. + + - `"response.mcp_call_arguments.done"` + +### Response Mcp Call Completed + +- `class ResponseMcpCallCompleted: …` + + Returned when an MCP tool call has completed successfully. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `type: Literal["response.mcp_call.completed"]` + + The event type, must be `response.mcp_call.completed`. + + - `"response.mcp_call.completed"` + +### Response Mcp Call Failed + +- `class ResponseMcpCallFailed: …` + + Returned when an MCP tool call has failed. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `type: Literal["response.mcp_call.failed"]` + + The event type, must be `response.mcp_call.failed`. + + - `"response.mcp_call.failed"` + +### Response Mcp Call In Progress + +- `class ResponseMcpCallInProgress: …` + + Returned when an MCP tool call has started and is in progress. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the MCP tool call item. + + - `output_index: int` + + The index of the output item in the response. + + - `type: Literal["response.mcp_call.in_progress"]` + + The event type, must be `response.mcp_call.in_progress`. + + - `"response.mcp_call.in_progress"` + +### Response Output Item Added Event + +- `class ResponseOutputItemAddedEvent: …` + + Returned when a new Item is created during Response generation. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `output_index: int` + + The index of the output item in the Response. + + - `response_id: str` + + The ID of the Response to which the item belongs. + + - `type: Literal["response.output_item.added"]` + + The event type, must be `response.output_item.added`. + + - `"response.output_item.added"` + +### Response Output Item Done Event + +- `class ResponseOutputItemDoneEvent: …` + + Returned when an Item is done streaming. Also emitted when a Response is + interrupted, incomplete, or cancelled. + + - `event_id: str` + + The unique ID of the server event. + + - `item: ConversationItem` + + A single item within a Realtime conversation. + + - `class RealtimeConversationItemSystemMessage: …` + + A system message in a Realtime conversation can be used to provide additional context or instructions to the model. This is similar but distinct from the instruction prompt provided at the start of a conversation, as system messages can be added at any point in the conversation. For major changes to the conversation's behavior, use instructions, but for smaller updates (e.g. "the user is now asking about a different topic"), use system messages. + + - `content: List[Content]` + + The content of the message. + + - `text: Optional[str]` + + The text content. + + - `type: Optional[Literal["input_text"]]` + + The content type. Always `input_text` for system messages. + + - `"input_text"` + + - `role: Literal["system"]` + + The role of the message sender. Always `system`. + + - `"system"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemUserMessage: …` + + A user message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `detail: Optional[Literal["auto", "low", "high"]]` + + The detail level of the image (for `input_image`). `auto` will default to `high`. + + - `"auto"` + + - `"low"` + + - `"high"` + + - `image_url: Optional[str]` + + Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG. + + - `text: Optional[str]` + + The text content (for `input_text`). + + - `transcript: Optional[str]` + + Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference. + + - `type: Optional[Literal["input_text", "input_audio", "input_image"]]` + + The content type (`input_text`, `input_audio`, or `input_image`). + + - `"input_text"` + + - `"input_audio"` + + - `"input_image"` + + - `role: Literal["user"]` + + The role of the message sender. Always `user`. + + - `"user"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemAssistantMessage: …` + + An assistant message item in a Realtime conversation. + + - `content: List[Content]` + + The content of the message. + + - `audio: Optional[str]` + + Base64-encoded audio bytes, these will be parsed as the format specified in the session output audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified. + + - `text: Optional[str]` + + The text content. + + - `transcript: Optional[str]` + + The transcript of the audio content, this will always be present if the output type is `audio`. + + - `type: Optional[Literal["output_text", "output_audio"]]` + + The content type, `output_text` or `output_audio` depending on the session `output_modalities` configuration. + + - `"output_text"` + + - `"output_audio"` + + - `role: Literal["assistant"]` + + The role of the message sender. Always `assistant`. + + - `"assistant"` + + - `type: Literal["message"]` + + The type of the item. Always `message`. + + - `"message"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCall: …` + + A function call item in a Realtime conversation. + + - `arguments: str` + + The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{"arg1": "value1", "arg2": 42}`. + + - `name: str` + + The name of the function being called. + + - `type: Literal["function_call"]` + + The type of the item. Always `function_call`. + + - `"function_call"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `call_id: Optional[str]` + + The ID of the function call. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeConversationItemFunctionCallOutput: …` + + A function call output item in a Realtime conversation. + + - `call_id: str` + + The ID of the function call this output is for. + + - `output: str` + + The output of the function call, this is free text and can contain any information or simply be empty. + + - `type: Literal["function_call_output"]` + + The type of the item. Always `function_call_output`. + + - `"function_call_output"` + + - `id: Optional[str]` + + The unique ID of the item. This may be provided by the client or generated by the server. + + - `object: Optional[Literal["realtime.item"]]` + + Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item. + + - `"realtime.item"` + + - `status: Optional[Literal["completed", "incomplete", "in_progress"]]` + + The status of the item. Has no effect on the conversation. + + - `"completed"` + + - `"incomplete"` + + - `"in_progress"` + + - `class RealtimeMcpApprovalResponse: …` + + A Realtime item responding to an MCP approval request. + + - `id: str` + + The unique ID of the approval response. + + - `approval_request_id: str` + + The ID of the approval request being answered. + + - `approve: bool` + + Whether the request was approved. + + - `type: Literal["mcp_approval_response"]` + + The type of the item. Always `mcp_approval_response`. + + - `"mcp_approval_response"` + + - `reason: Optional[str]` + + Optional reason for the decision. + + - `class RealtimeMcpListTools: …` + + A Realtime item listing tools available on an MCP server. + + - `server_label: str` + + The label of the MCP server. + + - `tools: List[Tool]` + + The tools available on the server. + + - `input_schema: object` + + The JSON schema describing the tool's input. + + - `name: str` + + The name of the tool. + + - `annotations: Optional[object]` + + Additional annotations about the tool. + + - `description: Optional[str]` + + The description of the tool. + + - `type: Literal["mcp_list_tools"]` + + The type of the item. Always `mcp_list_tools`. + + - `"mcp_list_tools"` + + - `id: Optional[str]` + + The unique ID of the list. + + - `class RealtimeMcpToolCall: …` + + A Realtime item representing an invocation of a tool on an MCP server. + + - `id: str` + + The unique ID of the tool call. + + - `arguments: str` + + A JSON string of the arguments passed to the tool. + + - `name: str` + + The name of the tool that was run. + + - `server_label: str` + + The label of the MCP server running the tool. + + - `type: Literal["mcp_call"]` + + The type of the item. Always `mcp_call`. + + - `"mcp_call"` + + - `approval_request_id: Optional[str]` + + The ID of an associated approval request, if any. + + - `error: Optional[Error]` + + The error from the tool call, if any. + + - `class RealtimeMcpProtocolError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["protocol_error"]` + + - `"protocol_error"` + + - `class RealtimeMcpToolExecutionError: …` + + - `message: str` + + - `type: Literal["tool_execution_error"]` + + - `"tool_execution_error"` + + - `class RealtimeMcphttpError: …` + + - `code: int` + + - `message: str` + + - `type: Literal["http_error"]` + + - `"http_error"` + + - `output: Optional[str]` + + The output from the tool call. + + - `class RealtimeMcpApprovalRequest: …` + + A Realtime item requesting human approval of a tool invocation. + + - `id: str` + + The unique ID of the approval request. + + - `arguments: str` + + A JSON string of arguments for the tool. + + - `name: str` + + The name of the tool to run. + + - `server_label: str` + + The label of the MCP server making the request. + + - `type: Literal["mcp_approval_request"]` + + The type of the item. Always `mcp_approval_request`. + + - `"mcp_approval_request"` + + - `output_index: int` + + The index of the output item in the Response. + + - `response_id: str` + + The ID of the Response to which the item belongs. + + - `type: Literal["response.output_item.done"]` + + The event type, must be `response.output_item.done`. + + - `"response.output_item.done"` + +### Response Text Delta Event + +- `class ResponseTextDeltaEvent: …` + + Returned when the text value of an "output_text" content part is updated. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `delta: str` + + The text delta. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `type: Literal["response.output_text.delta"]` + + The event type, must be `response.output_text.delta`. + + - `"response.output_text.delta"` + +### Response Text Done Event + +- `class ResponseTextDoneEvent: …` + + Returned when the text value of an "output_text" content part is done streaming. Also + emitted when a Response is interrupted, incomplete, or cancelled. + + - `content_index: int` + + The index of the content part in the item's content array. + + - `event_id: str` + + The unique ID of the server event. + + - `item_id: str` + + The ID of the item. + + - `output_index: int` + + The index of the output item in the response. + + - `response_id: str` + + The ID of the response. + + - `text: str` + + The final text content. + + - `type: Literal["response.output_text.done"]` + + The event type, must be `response.output_text.done`. + + - `"response.output_text.done"` + +### Session Created Event + +- `class SessionCreatedEvent: …` + + Returned when a Session is created. Emitted automatically when a new + connection is established as the first server event. This event will contain + the default Session configuration. + + - `event_id: str` + + The unique ID of the server event. + + - `session: Session` + + The session configuration. + + - `class RealtimeSessionCreateRequest: …` + + Realtime session object configuration. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[RealtimeAudioConfig]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[RealtimeToolChoiceConfig]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[RealtimeToolsConfig]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + + - `class RealtimeTranscriptionSessionCreateRequest: …` + + Realtime transcription session object configuration. + + - `type: Literal["transcription"]` + + The type of session to create. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[RealtimeTranscriptionSessionAudio]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeTranscriptionSessionAudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `type: Literal["session.created"]` + + The event type, must be `session.created`. + + - `"session.created"` + +### Session Update Event + +- `class SessionUpdateEvent: …` + + Send this event to update the session’s configuration. + The client may send this event at any time to update any field + except for `voice` and `model`. `voice` can be updated only if there have been no other audio outputs yet. + + When the server receives a `session.update`, it will respond + with a `session.updated` event showing the full, effective configuration. + Only the fields that are present in the `session.update` are updated. To clear a field like + `instructions`, pass an empty string. To clear a field like `tools`, pass an empty array. + To clear a field like `turn_detection`, pass `null`. + + - `session: Session` + + Update the Realtime session. Choose either a realtime + session or a transcription session. + + - `class RealtimeSessionCreateRequest: …` + + Realtime session object configuration. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[RealtimeAudioConfig]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[RealtimeToolChoiceConfig]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[RealtimeToolsConfig]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + + - `class RealtimeTranscriptionSessionCreateRequest: …` + + Realtime transcription session object configuration. + + - `type: Literal["transcription"]` + + The type of session to create. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[RealtimeTranscriptionSessionAudio]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeTranscriptionSessionAudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `type: Literal["session.update"]` + + The event type, must be `session.update`. + + - `"session.update"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. This is an arbitrary string that a client may assign. It will be passed back if there is an error with the event, but the corresponding `session.updated` event will not include it. + +### Session Updated Event + +- `class SessionUpdatedEvent: …` + + Returned when a session is updated with a `session.update` event, unless + there is an error. + + - `event_id: str` + + The unique ID of the server event. + + - `session: Session` + + The session configuration. + + - `class RealtimeSessionCreateRequest: …` + + Realtime session object configuration. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[RealtimeAudioConfig]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[RealtimeToolChoiceConfig]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[RealtimeToolsConfig]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + + - `class RealtimeTranscriptionSessionCreateRequest: …` + + Realtime transcription session object configuration. + + - `type: Literal["transcription"]` + + The type of session to create. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[RealtimeTranscriptionSessionAudio]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeTranscriptionSessionAudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `type: Literal["session.updated"]` + + The event type, must be `session.updated`. + + - `"session.updated"` + +### Transcription Session Update + +- `class TranscriptionSessionUpdate: …` + + Send this event to update a transcription session. + + - `session: Session` + + Realtime transcription session object configuration. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + The set of items to include in the transcription. Current available items are: + `item.input_audio_transcription.logprobs` + + - `"item.input_audio_transcription.logprobs"` + + - `input_audio_format: Optional[Literal["pcm16", "g711_ulaw", "g711_alaw"]]` + + The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. + For `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate, + single channel (mono), and little-endian byte order. + + - `"pcm16"` + + - `"g711_ulaw"` + + - `"g711_alaw"` + + - `input_audio_noise_reduction: Optional[SessionInputAudioNoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `input_audio_transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[SessionTurnDetection]` + + Configuration for turn detection. Can be set to `null` to turn off. Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + - `prefix_padding_ms: Optional[int]` + + Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `type: Optional[Literal["server_vad"]]` + + Type of turn detection. Only `server_vad` is currently supported for transcription sessions. + + - `"server_vad"` + + - `type: Literal["transcription_session.update"]` + + The event type, must be `transcription_session.update`. + + - `"transcription_session.update"` + + - `event_id: Optional[str]` + + Optional client-generated ID used to identify this event. + +### Transcription Session Updated Event + +- `class TranscriptionSessionUpdatedEvent: …` + + Returned when a transcription session is updated with a `transcription_session.update` event, unless + there is an error. + + - `event_id: str` + + The unique ID of the server event. + + - `session: Session` + + A new Realtime transcription session configuration. + + When a session is created on the server via REST API, the session object + also contains an ephemeral key. Default TTL for keys is 10 minutes. This + property is not present when a session is updated via the WebSocket API. + + - `client_secret: SessionClientSecret` + + Ephemeral key returned by the API. Only present when the session is + created on the server via REST API. + + - `expires_at: int` + + Timestamp for when the token expires. Currently, all tokens expire + after one minute. + + - `value: str` + + Ephemeral key usable in client environments to authenticate connections + to the Realtime API. Use this in client-side environments rather than + a standard API token, which should only be used server-side. + + - `input_audio_format: Optional[str]` + + The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. + + - `input_audio_transcription: Optional[AudioTranscription]` + + Configuration of the transcription model. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. To disable audio, + set this to ["text"]. + + - `"text"` + + - `"audio"` + + - `turn_detection: Optional[SessionTurnDetection]` + + Configuration for turn detection. Can be set to `null` to turn off. Server + VAD means that the model will detect the start and end of speech based on + audio volume and respond at the end of user speech. + + - `prefix_padding_ms: Optional[int]` + + Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `type: Optional[str]` + + Type of turn detection, only `server_vad` is currently supported. + + - `type: Literal["transcription_session.updated"]` + + The event type, must be `transcription_session.updated`. + + - `"transcription_session.updated"` + +# Client Secrets + +## Create client secret + +`realtime.client_secrets.create(ClientSecretCreateParams**kwargs) -> ClientSecretCreateResponse` + +**post** `/realtime/client_secrets` + +Create a Realtime client secret with an associated session configuration. + +Client secrets are short-lived tokens that can be passed to a client app, +such as a web frontend or mobile client, which grants access to the Realtime API without +leaking your main API key. You can configure a custom TTL for each client secret. + +You can also attach session configuration options to the client secret, which will be +applied to any sessions created using that client secret, but these can also be overridden +by the client connection. + +[Learn more about authentication with client secrets over WebRTC](https://platform.openai.com/docs/guides/realtime-webrtc). + +Returns the created client secret and the effective session object. The client secret is a string that looks like `ek_1234`. + +### Parameters + +- `expires_after: Optional[ExpiresAfter]` + + Configuration for the client secret expiration. Expiration refers to the time after which + a client secret will no longer be valid for creating sessions. The session itself may + continue after that time once started. A secret can be used to create multiple sessions + until it expires. + + - `anchor: Optional[Literal["created_at"]]` + + The anchor point for the client secret expiration, meaning that `seconds` will be added to the `created_at` time of the client secret to produce an expiration timestamp. Only `created_at` is currently supported. + + - `"created_at"` + + - `seconds: Optional[int]` + + The number of seconds from the anchor point to the expiration. Select a value between `10` and `7200` (2 hours). This default to 600 seconds (10 minutes) if not specified. + +- `session: Optional[Session]` + + Session configuration to use for the client secret. Choose either a realtime + session or a transcription session. + + - `class RealtimeSessionCreateRequest: …` + + Realtime session object configuration. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[RealtimeAudioConfig]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[RealtimeToolChoiceConfig]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[RealtimeToolsConfig]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + + - `class RealtimeTranscriptionSessionCreateRequest: …` + + Realtime transcription session object configuration. + + - `type: Literal["transcription"]` + + The type of session to create. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[RealtimeTranscriptionSessionAudio]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeTranscriptionSessionAudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `turn_detection: Optional[RealtimeTranscriptionSessionAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + +### Returns + +- `class ClientSecretCreateResponse: …` + + Response from creating a session and client secret for the Realtime API. + + - `expires_at: int` + + Expiration timestamp for the client secret, in seconds since epoch. + + - `session: Session` + + The session configuration for either a realtime or transcription session. + + - `class RealtimeSessionCreateResponse: …` + + A new Realtime session configuration, with an ephemeral key. Default TTL + for keys is one minute. + + - `client_secret: RealtimeSessionClientSecret` + + Ephemeral key returned by the API. + + - `expires_at: int` + + Timestamp for when the token expires. Currently, all tokens expire + after one minute. + + - `value: str` + + Ephemeral key usable in client environments to authenticate connections to the Realtime API. Use this in client-side environments rather than a standard API token, which should only be used server-side. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[Audio]` + + Configuration for input and output audio. + + - `input: Optional[AudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[AudioInputNoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[AudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class AudioInputTurnDetectionServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class AudioInputTurnDetectionSemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[AudioOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Union[str, Literal["alloy", "ash", "ballad", 7 more], null]]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[ToolChoice]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[List[Tool]]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class ToolMcpTool: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[ToolMcpToolAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class ToolMcpToolAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[ToolMcpToolRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class ToolMcpToolRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[ToolMcpToolRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[ToolMcpToolRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[Tracing]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingTracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + + - `class RealtimeTranscriptionSessionCreateResponse: …` + + A Realtime transcription session configuration object. + + - `id: str` + + Unique identifier for the session that looks like `sess_1234567890abcdef`. + + - `object: str` + + The object type. Always `realtime.transcription_session`. + + - `type: Literal["transcription"]` + + The type of session. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[Audio]` + + Configuration for input audio for the session. + + - `input: Optional[AudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `noise_reduction: Optional[AudioInputNoiseReduction]` + + Configuration for input audio noise reduction. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `transcription: Optional[AudioTranscription]` + + Configuration of the transcription model. + + - `turn_detection: Optional[RealtimeTranscriptionSessionTurnDetection]` + + Configuration for turn detection. Can be set to `null` to turn off. Server + VAD means that the model will detect the start and end of speech based on + audio volume and respond at the end of user speech. + + - `prefix_padding_ms: Optional[int]` + + Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `type: Optional[str]` + + Type of turn detection, only `server_vad` is currently supported. + + - `expires_at: Optional[int]` + + Expiration timestamp for the session, in seconds since epoch. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + - `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `value: str` + + The generated client secret value. + +### Example + +```python +import os +from openai import OpenAI + +client = OpenAI( + api_key=os.environ.get("OPENAI_API_KEY"), # This is the default and can be omitted +) +client_secret = client.realtime.client_secrets.create() +print(client_secret.expires_at) +``` + +#### Response + +```json +{ + "expires_at": 0, + "session": { + "client_secret": { + "expires_at": 0, + "value": "value" + }, + "type": "realtime", + "audio": { + "input": { + "format": { + "rate": 24000, + "type": "audio/pcm" + }, + "noise_reduction": { + "type": "near_field" + }, + "transcription": { + "language": "language", + "model": "string", + "prompt": "prompt" + }, + "turn_detection": { + "type": "server_vad", + "create_response": true, + "idle_timeout_ms": 5000, + "interrupt_response": true, + "prefix_padding_ms": 0, + "silence_duration_ms": 0, + "threshold": 0 + } + }, + "output": { + "format": { + "rate": 24000, + "type": "audio/pcm" + }, + "speed": 0.25, + "voice": "ash" + } + }, + "include": [ + "item.input_audio_transcription.logprobs" + ], + "instructions": "instructions", + "max_output_tokens": 0, + "model": "string", + "output_modalities": [ + "text" + ], + "prompt": { + "id": "id", + "variables": { + "foo": "string" + }, + "version": "version" + }, + "tool_choice": "none", + "tools": [ + { + "description": "description", + "name": "name", + "parameters": {}, + "type": "function" + } + ], + "tracing": "auto", + "truncation": "auto" + }, + "value": "value" +} +``` + +## Domain Types + +### Realtime Session Client Secret + +- `class RealtimeSessionClientSecret: …` + + Ephemeral key returned by the API. + + - `expires_at: int` + + Timestamp for when the token expires. Currently, all tokens expire + after one minute. + + - `value: str` + + Ephemeral key usable in client environments to authenticate connections to the Realtime API. Use this in client-side environments rather than a standard API token, which should only be used server-side. + +### Realtime Session Create Response + +- `class RealtimeSessionCreateResponse: …` + + A new Realtime session configuration, with an ephemeral key. Default TTL + for keys is one minute. + + - `client_secret: RealtimeSessionClientSecret` + + Ephemeral key returned by the API. + + - `expires_at: int` + + Timestamp for when the token expires. Currently, all tokens expire + after one minute. + + - `value: str` + + Ephemeral key usable in client environments to authenticate connections to the Realtime API. Use this in client-side environments rather than a standard API token, which should only be used server-side. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[Audio]` + + Configuration for input and output audio. + + - `input: Optional[AudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[AudioInputNoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[AudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class AudioInputTurnDetectionServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class AudioInputTurnDetectionSemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[AudioOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Union[str, Literal["alloy", "ash", "ballad", 7 more], null]]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[ToolChoice]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[List[Tool]]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class ToolMcpTool: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[ToolMcpToolAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class ToolMcpToolAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[ToolMcpToolRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class ToolMcpToolRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[ToolMcpToolRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[ToolMcpToolRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[Tracing]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingTracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + +### Realtime Transcription Session Create Response + +- `class RealtimeTranscriptionSessionCreateResponse: …` + + A Realtime transcription session configuration object. + + - `id: str` + + Unique identifier for the session that looks like `sess_1234567890abcdef`. + + - `object: str` + + The object type. Always `realtime.transcription_session`. + + - `type: Literal["transcription"]` + + The type of session. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[Audio]` + + Configuration for input audio for the session. + + - `input: Optional[AudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[AudioInputNoiseReduction]` + + Configuration for input audio noise reduction. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration of the transcription model. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeTranscriptionSessionTurnDetection]` + + Configuration for turn detection. Can be set to `null` to turn off. Server + VAD means that the model will detect the start and end of speech based on + audio volume and respond at the end of user speech. + + - `prefix_padding_ms: Optional[int]` + + Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `type: Optional[str]` + + Type of turn detection, only `server_vad` is currently supported. + + - `expires_at: Optional[int]` + + Expiration timestamp for the session, in seconds since epoch. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + - `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + +### Realtime Transcription Session Turn Detection + +- `class RealtimeTranscriptionSessionTurnDetection: …` + + Configuration for turn detection. Can be set to `null` to turn off. Server + VAD means that the model will detect the start and end of speech based on + audio volume and respond at the end of user speech. + + - `prefix_padding_ms: Optional[int]` + + Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `type: Optional[str]` + + Type of turn detection, only `server_vad` is currently supported. + +### Client Secret Create Response + +- `class ClientSecretCreateResponse: …` + + Response from creating a session and client secret for the Realtime API. + + - `expires_at: int` + + Expiration timestamp for the client secret, in seconds since epoch. + + - `session: Session` + + The session configuration for either a realtime or transcription session. + + - `class RealtimeSessionCreateResponse: …` + + A new Realtime session configuration, with an ephemeral key. Default TTL + for keys is one minute. + + - `client_secret: RealtimeSessionClientSecret` + + Ephemeral key returned by the API. + + - `expires_at: int` + + Timestamp for when the token expires. Currently, all tokens expire + after one minute. + + - `value: str` + + Ephemeral key usable in client environments to authenticate connections to the Realtime API. Use this in client-side environments rather than a standard API token, which should only be used server-side. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[Audio]` + + Configuration for input and output audio. + + - `input: Optional[AudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[AudioInputNoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[AudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class AudioInputTurnDetectionServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class AudioInputTurnDetectionSemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[AudioOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Union[str, Literal["alloy", "ash", "ballad", 7 more], null]]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + The voice the model uses to respond. Voice cannot be changed during the + session once the model has responded with audio at least once. Current + voice options are `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, + `shimmer`, `verse`, `marin`, and `cedar`. We recommend `marin` and `cedar` for + best quality. + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[ToolChoice]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[List[Tool]]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class ToolMcpTool: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[ToolMcpToolAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class ToolMcpToolAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[ToolMcpToolRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class ToolMcpToolRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[ToolMcpToolRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[ToolMcpToolRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[Tracing]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingTracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + + - `class RealtimeTranscriptionSessionCreateResponse: …` + + A Realtime transcription session configuration object. + + - `id: str` + + Unique identifier for the session that looks like `sess_1234567890abcdef`. + + - `object: str` + + The object type. Always `realtime.transcription_session`. + + - `type: Literal["transcription"]` + + The type of session. Always `transcription` for transcription sessions. + + - `"transcription"` + + - `audio: Optional[Audio]` + + Configuration for input audio for the session. + + - `input: Optional[AudioInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `noise_reduction: Optional[AudioInputNoiseReduction]` + + Configuration for input audio noise reduction. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `transcription: Optional[AudioTranscription]` + + Configuration of the transcription model. + + - `turn_detection: Optional[RealtimeTranscriptionSessionTurnDetection]` + + Configuration for turn detection. Can be set to `null` to turn off. Server + VAD means that the model will detect the start and end of speech based on + audio volume and respond at the end of user speech. + + - `prefix_padding_ms: Optional[int]` + + Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `type: Optional[str]` + + Type of turn detection, only `server_vad` is currently supported. + + - `expires_at: Optional[int]` + + Expiration timestamp for the session, in seconds since epoch. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + - `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `value: str` + + The generated client secret value. + +# Calls + +## Accept call + +`realtime.calls.accept(strcall_id, CallAcceptParams**kwargs)` + +**post** `/realtime/calls/{call_id}/accept` + +Accept an incoming SIP call and configure the realtime session that will +handle it. + +### Parameters + +- `call_id: str` + +- `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + +- `audio: Optional[RealtimeAudioConfigParam]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + +- `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + +- `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + +- `max_output_tokens: Optional[Union[int, Literal["inf"]]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + +- `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + +- `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + +- `prompt: Optional[ResponsePromptParam]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + +- `tool_choice: Optional[RealtimeToolChoiceConfigParam]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + +- `tools: Optional[RealtimeToolsConfigParam]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + +- `tracing: Optional[RealtimeTracingConfigParam]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + +- `truncation: Optional[RealtimeTruncationParam]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + +### Example + +```python +import os +from openai import OpenAI + +client = OpenAI( + api_key=os.environ.get("OPENAI_API_KEY"), # This is the default and can be omitted +) +client.realtime.calls.accept( + call_id="call_id", + type="realtime", +) +``` + +## Hang up call + +`realtime.calls.hangup(strcall_id)` + +**post** `/realtime/calls/{call_id}/hangup` + +End an active Realtime API call, whether it was initiated over SIP or +WebRTC. + +### Parameters + +- `call_id: str` + +### Example + +```python +import os +from openai import OpenAI + +client = OpenAI( + api_key=os.environ.get("OPENAI_API_KEY"), # This is the default and can be omitted +) +client.realtime.calls.hangup( + "call_id", +) +``` + +## Refer call + +`realtime.calls.refer(strcall_id, CallReferParams**kwargs)` + +**post** `/realtime/calls/{call_id}/refer` + +Transfer an active SIP call to a new destination using the SIP REFER verb. + +### Parameters + +- `call_id: str` + +- `target_uri: str` + + URI that should appear in the SIP Refer-To header. Supports values like + `tel:+14155550123` or `sip:agent@example.com`. + +### Example + +```python +import os +from openai import OpenAI + +client = OpenAI( + api_key=os.environ.get("OPENAI_API_KEY"), # This is the default and can be omitted +) +client.realtime.calls.refer( + call_id="call_id", + target_uri="tel:+14155550123", +) +``` + +## Reject call + +`realtime.calls.reject(strcall_id, CallRejectParams**kwargs)` + +**post** `/realtime/calls/{call_id}/reject` + +Decline an incoming SIP call by returning a SIP status code to the caller. + +### Parameters + +- `call_id: str` + +- `status_code: Optional[int]` + + SIP response code to send back to the caller. Defaults to `603` (Decline) + when omitted. + +### Example + +```python +import os +from openai import OpenAI + +client = OpenAI( + api_key=os.environ.get("OPENAI_API_KEY"), # This is the default and can be omitted +) +client.realtime.calls.reject( + call_id="call_id", +) +``` + +## Create call + +`realtime.calls.create(CallCreateParams**kwargs) -> BinaryResponseContent` + +**post** `/realtime/calls` + +Create a new Realtime API call over WebRTC and receive the SDP answer needed +to complete the peer connection. + +### Parameters + +- `sdp: str` + + WebRTC Session Description Protocol (SDP) offer generated by the caller. + +- `session: Optional[RealtimeSessionCreateRequestParam]` + + Realtime session object configuration. + + - `type: Literal["realtime"]` + + The type of session to create. Always `realtime` for the Realtime API. + + - `"realtime"` + + - `audio: Optional[RealtimeAudioConfig]` + + Configuration for input and output audio. + + - `input: Optional[RealtimeAudioConfigInput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the input audio. + + - `class AudioPCM: …` + + The PCM audio format. Only a 24kHz sample rate is supported. + + - `rate: Optional[Literal[24000]]` + + The sample rate of the audio. Always `24000`. + + - `24000` + + - `type: Optional[Literal["audio/pcm"]]` + + The audio format. Always `audio/pcm`. + + - `"audio/pcm"` + + - `class AudioPCMU: …` + + The G.711 μ-law format. + + - `type: Optional[Literal["audio/pcmu"]]` + + The audio format. Always `audio/pcmu`. + + - `"audio/pcmu"` + + - `class AudioPCMA: …` + + The G.711 A-law format. + + - `type: Optional[Literal["audio/pcma"]]` + + The audio format. Always `audio/pcma`. + + - `"audio/pcma"` + + - `noise_reduction: Optional[NoiseReduction]` + + Configuration for input audio noise reduction. This can be set to `null` to turn off. + Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model. + Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio. + + - `type: Optional[NoiseReductionType]` + + Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones. + + - `"near_field"` + + - `"far_field"` + + - `transcription: Optional[AudioTranscription]` + + Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service. + + - `language: Optional[str]` + + The language of the input audio. Supplying the input language in + [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format + will improve accuracy and latency. + + - `model: Optional[Union[str, Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more], null]]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `str` + + - `Literal["whisper-1", "gpt-4o-mini-transcribe", "gpt-4o-mini-transcribe-2025-12-15", 2 more]` + + The model to use for transcription. Current options are `whisper-1`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels. + + - `"whisper-1"` + + - `"gpt-4o-mini-transcribe"` + + - `"gpt-4o-mini-transcribe-2025-12-15"` + + - `"gpt-4o-transcribe"` + + - `"gpt-4o-transcribe-diarize"` + + - `prompt: Optional[str]` + + An optional text to guide the model's style or continue a previous audio + segment. + For `whisper-1`, the [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting). + For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example "expect words related to technology". + + - `turn_detection: Optional[RealtimeAudioInputTurnDetection]` + + Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to `null` to turn off, in which case the client must manually trigger model response. + + Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. + + Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. + + - `class ServerVad: …` + + Server-side voice activity detection (VAD) which flips on when user speech is detected and off after a period of silence. + + - `type: Literal["server_vad"]` + + Type of turn detection, `server_vad` to turn on simple Server VAD. + + - `"server_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `idle_timeout_ms: Optional[int]` + + Optional timeout after which a model response will be triggered automatically. This is + useful for situations in which a long pause from the user is unexpected, such as a phone + call. The model will effectively prompt the user to continue the conversation based + on the current context. + + The timeout value will be applied after the last model response's audio has finished playing, + i.e. it's set to the `response.done` time plus audio playback duration. + + An `input_audio_buffer.timeout_triggered` event (plus events + associated with the Response) will be emitted when the timeout is reached. + Idle timeout is currently only supported for `server_vad` mode. + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt (cancel) any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete. + + If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted. + + - `prefix_padding_ms: Optional[int]` + + Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in + milliseconds). Defaults to 300ms. + + - `silence_duration_ms: Optional[int]` + + Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults + to 500ms. With shorter values the model will respond more quickly, + but may jump in on short pauses from the user. + + - `threshold: Optional[float]` + + Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A + higher threshold will require louder audio to activate the model, and + thus might perform better in noisy environments. + + - `class SemanticVad: …` + + Server-side semantic turn detection which uses a model to determine when the user has finished speaking. + + - `type: Literal["semantic_vad"]` + + Type of turn detection, `semantic_vad` to turn on Semantic VAD. + + - `"semantic_vad"` + + - `create_response: Optional[bool]` + + Whether or not to automatically generate a response when a VAD stop event occurs. + + - `eagerness: Optional[Literal["low", "medium", "high", "auto"]]` + + Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively. + + - `"low"` + + - `"medium"` + + - `"high"` + + - `"auto"` + + - `interrupt_response: Optional[bool]` + + Whether or not to automatically interrupt any ongoing response with output to the default + conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. + + - `output: Optional[RealtimeAudioConfigOutput]` + + - `format: Optional[RealtimeAudioFormats]` + + The format of the output audio. + + - `speed: Optional[float]` + + The speed of the model's spoken response as a multiple of the original speed. + 1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress. + + This parameter is a post-processing adjustment to the audio after it is generated, it's + also possible to prompt the model to speak faster or slower. + + - `voice: Optional[Voice]` + + The voice the model uses to respond. Supported built-in voices are + `alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, + `marin`, and `cedar`. You may also provide a custom voice object with + an `id`, for example `{ "id": "voice_1234" }`. Voice cannot be changed + during the session once the model has responded with audio at least once. + We recommend `marin` and `cedar` for best quality. + + - `str` + + - `Literal["alloy", "ash", "ballad", 7 more]` + + - `"alloy"` + + - `"ash"` + + - `"ballad"` + + - `"coral"` + + - `"echo"` + + - `"sage"` + + - `"shimmer"` + + - `"verse"` + + - `"marin"` + + - `"cedar"` + + - `class VoiceID: …` + + Custom voice reference. + + - `id: str` + + The custom voice ID, e.g. `voice_1234`. + + - `include: Optional[List[Literal["item.input_audio_transcription.logprobs"]]]` + + Additional fields to include in server outputs. + + `item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription. + + - `"item.input_audio_transcription.logprobs"` + + - `instructions: Optional[str]` + + The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. + + Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session. + + - `max_output_tokens: Optional[Union[int, Literal["inf"], null]]` + + Maximum number of output tokens for a single assistant response, + inclusive of tool calls. Provide an integer between 1 and 4096 to + limit output tokens, or `inf` for the maximum available tokens for a + given model. Defaults to `inf`. + + - `int` + + - `Literal["inf"]` + + - `"inf"` + + - `model: Optional[Union[str, Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more], null]]` + + The Realtime model used for this session. + + - `str` + + - `Literal["gpt-realtime", "gpt-realtime-1.5", "gpt-realtime-2025-08-28", 13 more]` + + The Realtime model used for this session. + + - `"gpt-realtime"` + + - `"gpt-realtime-1.5"` + + - `"gpt-realtime-2025-08-28"` + + - `"gpt-4o-realtime-preview"` + + - `"gpt-4o-realtime-preview-2024-10-01"` + + - `"gpt-4o-realtime-preview-2024-12-17"` + + - `"gpt-4o-realtime-preview-2025-06-03"` + + - `"gpt-4o-mini-realtime-preview"` + + - `"gpt-4o-mini-realtime-preview-2024-12-17"` + + - `"gpt-realtime-mini"` + + - `"gpt-realtime-mini-2025-10-06"` + + - `"gpt-realtime-mini-2025-12-15"` + + - `"gpt-audio-1.5"` + + - `"gpt-audio-mini"` + + - `"gpt-audio-mini-2025-10-06"` + + - `"gpt-audio-mini-2025-12-15"` + + - `output_modalities: Optional[List[Literal["text", "audio"]]]` + + The set of modalities the model can respond with. It defaults to `["audio"]`, indicating + that the model will respond with audio plus a transcript. `["text"]` can be used to make + the model respond with text only. It is not possible to request both `text` and `audio` at the same time. + + - `"text"` + + - `"audio"` + + - `prompt: Optional[ResponsePrompt]` + + Reference to a prompt template and its variables. + [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts). + + - `id: str` + + The unique identifier of the prompt template to use. + + - `variables: Optional[Dict[str, Variables]]` + + Optional map of values to substitute in for variables in your + prompt. The substitution values can either be strings, or other + Response input types like images or files. + + - `str` + + - `class ResponseInputText: …` + + A text input to the model. + + - `text: str` + + The text input to the model. + + - `type: Literal["input_text"]` + + The type of the input item. Always `input_text`. + + - `"input_text"` + + - `class ResponseInputImage: …` + + An image input to the model. Learn about [image inputs](https://platform.openai.com/docs/guides/vision). + + - `detail: Literal["low", "high", "auto", "original"]` + + The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`. + + - `"low"` + + - `"high"` + + - `"auto"` + + - `"original"` + + - `type: Literal["input_image"]` + + The type of the input item. Always `input_image`. + + - `"input_image"` + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `image_url: Optional[str]` + + The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL. + + - `class ResponseInputFile: …` + + A file input to the model. + + - `type: Literal["input_file"]` + + The type of the input item. Always `input_file`. + + - `"input_file"` + + - `detail: Optional[Literal["low", "high"]]` + + The detail level of the file to be sent to the model. Use `low` for the default rendering behavior, or `high` to render the file at higher quality. Defaults to `low`. + + - `"low"` + + - `"high"` + + - `file_data: Optional[str]` + + The content of the file to be sent to the model. + + - `file_id: Optional[str]` + + The ID of the file to be sent to the model. + + - `file_url: Optional[str]` + + The URL of the file to be sent to the model. + + - `filename: Optional[str]` + + The name of the file to be sent to the model. + + - `version: Optional[str]` + + Optional version of the prompt template. + + - `tool_choice: Optional[RealtimeToolChoiceConfig]` + + How the model chooses tools. Provide one of the string modes or force a specific + function/MCP tool. + + - `Literal["none", "auto", "required"]` + + - `"none"` + + - `"auto"` + + - `"required"` + + - `class ToolChoiceFunction: …` + + Use this option to force the model to call a specific function. + + - `name: str` + + The name of the function to call. + + - `type: Literal["function"]` + + For function calling, the type is always `function`. + + - `"function"` + + - `class ToolChoiceMcp: …` + + Use this option to force the model to call a specific tool on a remote MCP server. + + - `server_label: str` + + The label of the MCP server to use. + + - `type: Literal["mcp"]` + + For MCP tools, the type is always `mcp`. + + - `"mcp"` + + - `name: Optional[str]` + + The name of the tool to call on the server. + + - `tools: Optional[RealtimeToolsConfig]` + + Tools available to the model. + + - `class RealtimeFunctionTool: …` + + - `description: Optional[str]` + + The description of the function, including guidance on when and how + to call it, and guidance about what to tell the user when calling + (if anything). + + - `name: Optional[str]` + + The name of the function. + + - `parameters: Optional[object]` + + Parameters of the function in JSON Schema. + + - `type: Optional[Literal["function"]]` + + The type of the tool, i.e. `function`. + + - `"function"` + + - `class Mcp: …` + + Give the model access to additional tools via remote Model Context Protocol + (MCP) servers. [Learn more about MCP](https://platform.openai.com/docs/guides/tools-remote-mcp). + + - `server_label: str` + + A label for this MCP server, used to identify it in tool calls. + + - `type: Literal["mcp"]` + + The type of the MCP tool. Always `mcp`. + + - `"mcp"` + + - `allowed_tools: Optional[McpAllowedTools]` + + List of allowed tool names or a filter object. + + - `List[str]` + + A string array of allowed tool names + + - `class McpAllowedToolsMcpToolFilter: …` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `authorization: Optional[str]` + + An OAuth access token that can be used with a remote MCP server, either + with a custom MCP server URL or a service connector. Your application + must handle the OAuth authorization flow and provide the token here. + + - `connector_id: Optional[Literal["connector_dropbox", "connector_gmail", "connector_googlecalendar", 5 more]]` + + Identifier for service connectors, like those available in ChatGPT. One of + `server_url` or `connector_id` must be provided. Learn more about service + connectors [here](https://platform.openai.com/docs/guides/tools-remote-mcp#connectors). + + Currently supported `connector_id` values are: + + - Dropbox: `connector_dropbox` + - Gmail: `connector_gmail` + - Google Calendar: `connector_googlecalendar` + - Google Drive: `connector_googledrive` + - Microsoft Teams: `connector_microsoftteams` + - Outlook Calendar: `connector_outlookcalendar` + - Outlook Email: `connector_outlookemail` + - SharePoint: `connector_sharepoint` + + - `"connector_dropbox"` + + - `"connector_gmail"` + + - `"connector_googlecalendar"` + + - `"connector_googledrive"` + + - `"connector_microsoftteams"` + + - `"connector_outlookcalendar"` + + - `"connector_outlookemail"` + + - `"connector_sharepoint"` + + - `defer_loading: Optional[bool]` + + Whether this MCP tool is deferred and discovered via tool search. + + - `headers: Optional[Dict[str, str]]` + + Optional HTTP headers to send to the MCP server. Use for authentication + or other purposes. + + - `require_approval: Optional[McpRequireApproval]` + + Specify which of the MCP server's tools require approval. + + - `class McpRequireApprovalMcpToolApprovalFilter: …` + + Specify which of the MCP server's tools require approval. Can be + `always`, `never`, or a filter object associated with tools + that require approval. + + - `always: Optional[McpRequireApprovalMcpToolApprovalFilterAlways]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `never: Optional[McpRequireApprovalMcpToolApprovalFilterNever]` + + A filter object to specify which tools are allowed. + + - `read_only: Optional[bool]` + + Indicates whether or not a tool modifies data or is read-only. If an + MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), + it will match this filter. + + - `tool_names: Optional[List[str]]` + + List of allowed tool names. + + - `Literal["always", "never"]` + + Specify a single approval policy for all tools. One of `always` or + `never`. When set to `always`, all tools will require approval. When + set to `never`, all tools will not require approval. + + - `"always"` + + - `"never"` + + - `server_description: Optional[str]` + + Optional description of the MCP server, used to provide more context. + + - `server_url: Optional[str]` + + The URL for the MCP server. One of `server_url` or `connector_id` must be + provided. + + - `tracing: Optional[RealtimeTracingConfig]` + + Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once + tracing is enabled for a session, the configuration cannot be modified. + + `auto` will create a trace for the session with default values for the + workflow name, group id, and metadata. + + - `Literal["auto"]` + + Enables tracing and sets default values for tracing configuration options. Always `auto`. + + - `"auto"` + + - `class TracingConfiguration: …` + + Granular configuration for tracing. + + - `group_id: Optional[str]` + + The group id to attach to this trace to enable filtering and + grouping in the Traces Dashboard. + + - `metadata: Optional[object]` + + The arbitrary metadata to attach to this trace to enable + filtering in the Traces Dashboard. + + - `workflow_name: Optional[str]` + + The name of the workflow to attach to this trace. This is used to + name the trace in the Traces Dashboard. + + - `truncation: Optional[RealtimeTruncation]` + + When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. + + Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. + + Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. + + Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit. + + - `Literal["auto", "disabled"]` + + The truncation strategy to use for the session. `auto` is the default truncation strategy. `disabled` will disable truncation and emit errors when the conversation exceeds the input token limit. + + - `"auto"` + + - `"disabled"` + + - `class RealtimeTruncationRetentionRatio: …` + + Retain a fraction of the conversation tokens when the conversation exceeds the input token limit. This allows you to amortize truncations across multiple turns, which can help improve cached token usage. + + - `retention_ratio: float` + + Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates. + + - `type: Literal["retention_ratio"]` + + Use retention ratio truncation. + + - `"retention_ratio"` + + - `token_limits: Optional[TokenLimits]` + + Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used. + + - `post_instructions: Optional[int]` + + Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens. + +### Returns + +- `BinaryResponseContent` + +### Example + +```python +import os +from openai import OpenAI + +client = OpenAI( + api_key=os.environ.get("OPENAI_API_KEY"), # This is the default and can be omitted +) +call = client.realtime.calls.create( + sdp="sdp", +) +print(call) +content = call.read() +print(content) +``` + +# Sessions + +# Transcription Sessions