Assistants

List assistants

beta.assistants.list(AssistantListParams**kwargs) -> SyncCursorPage[Assistant]

get /assistants

List assistants

Parameters

• after: Optional[str]

  A cursor for use in pagination. after is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.

• before: Optional[str]

  A cursor for use in pagination. before is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.

• limit: Optional[int]

  A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.

• order: Optional[Literal["asc", "desc"]]

  Sort order by the created_at timestamp of the objects. asc for ascending order and desc for descending order.

  • "asc"

  • "desc"

Returns

• class Assistant: …

  Represents an assistant that can call the model and use tools.

  • id: str

    The identifier, which can be referenced in API endpoints.

  • created_at: int

    The Unix timestamp (in seconds) for when the assistant was created.

  • description: Optional[str]

    The description of the assistant. The maximum length is 512 characters.

  • instructions: Optional[str]

    The system instructions that the assistant uses. The maximum length is 256,000 characters.

  • 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.

  • model: str

    ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.

  • name: Optional[str]

    The name of the assistant. The maximum length is 256 characters.

  • object: Literal["assistant"]

    The object type, which is always assistant.

    • "assistant"

  • tools: List[object]

    A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types code_interpreter, file_search, or function.

  • response_format: Optional[AssistantResponseFormatOption]

    Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

    Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

    Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

    Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

    • Literal["auto"]

      auto is the default value

      • "auto"

    • class ResponseFormatText: …

      Default response format. Used to generate text responses.

      • type: Literal["text"]

        The type of response format being defined. Always text.

        • "text"

    • class ResponseFormatJSONObject: …

      JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

      • type: Literal["json_object"]

        The type of response format being defined. Always json_object.

        • "json_object"

    • class ResponseFormatJSONSchema: …

      JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

      • json_schema: JSONSchema

        Structured Outputs configuration options, including a JSON Schema.

        • name: str

          The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

        • description: Optional[str]

          A description of what the response format is for, used by the model to determine how to respond in the format.

        • schema: Optional[Dict[str, object]]

          The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

        • strict: Optional[bool]

          Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

      • type: Literal["json_schema"]

        The type of response format being defined. Always json_schema.

        • "json_schema"

  • temperature: Optional[float]

    What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

  • tool_resources: Optional[ToolResources]

    A set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

    • code_interpreter: Optional[ToolResourcesCodeInterpreter]

      • file_ids: Optional[List[str]]

        A list of file IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

    • file_search: Optional[ToolResourcesFileSearch]

      • vector_store_ids: Optional[List[str]]

        The ID of the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

  • top_p: Optional[float]

    An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

    We generally recommend altering this or temperature but not both.

Example

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ.get("OPENAI_API_KEY"),  # This is the default and can be omitted
)
page = client.beta.assistants.list()
page = page.data[0]
print(page.id)

Response

{
  "data": [
    {
      "id": "id",
      "created_at": 0,
      "description": "description",
      "instructions": "instructions",
      "metadata": {
        "foo": "string"
      },
      "model": "model",
      "name": "name",
      "object": "assistant",
      "tools": [
        {}
      ],
      "response_format": "auto",
      "temperature": 1,
      "tool_resources": {
        "code_interpreter": {
          "file_ids": [
            "string"
          ]
        },
        "file_search": {
          "vector_store_ids": [
            "string"
          ]
        }
      },
      "top_p": 1
    }
  ],
  "first_id": "asst_abc123",
  "has_more": false,
  "last_id": "asst_abc456",
  "object": "list"
}

Example

from openai import OpenAI
client = OpenAI()

my_assistants = client.beta.assistants.list(
    order="desc",
    limit="20",
)
print(my_assistants.data)

Response

{
  "object": "list",
  "data": [
    {
      "id": "asst_abc123",
      "object": "assistant",
      "created_at": 1698982736,
      "name": "Coding Tutor",
      "description": null,
      "model": "gpt-4o",
      "instructions": "You are a helpful assistant designed to make me better at coding!",
      "tools": [],
      "tool_resources": {},
      "metadata": {},
      "top_p": 1.0,
      "temperature": 1.0,
      "response_format": "auto"
    },
    {
      "id": "asst_abc456",
      "object": "assistant",
      "created_at": 1698982718,
      "name": "My Assistant",
      "description": null,
      "model": "gpt-4o",
      "instructions": "You are a helpful assistant designed to make me better at coding!",
      "tools": [],
      "tool_resources": {},
      "metadata": {},
      "top_p": 1.0,
      "temperature": 1.0,
      "response_format": "auto"
    },
    {
      "id": "asst_abc789",
      "object": "assistant",
      "created_at": 1698982643,
      "name": null,
      "description": null,
      "model": "gpt-4o",
      "instructions": null,
      "tools": [],
      "tool_resources": {},
      "metadata": {},
      "top_p": 1.0,
      "temperature": 1.0,
      "response_format": "auto"
    }
  ],
  "first_id": "asst_abc123",
  "last_id": "asst_abc789",
  "has_more": false
}

Create assistant

beta.assistants.create(AssistantCreateParams**kwargs) -> Assistant

post /assistants

Create assistant

Parameters

• model: Union[str, ChatModel]

  ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.

  • str

  • Literal["gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna", 78 more]

    • "gpt-5.6-sol"

    • "gpt-5.6-terra"

    • "gpt-5.6-luna"

    • "gpt-5.4"

    • "gpt-5.4-mini"

    • "gpt-5.4-nano"

    • "gpt-5.4-mini-2026-03-17"

    • "gpt-5.4-nano-2026-03-17"

    • "gpt-5.3-chat-latest"

    • "gpt-5.2"

    • "gpt-5.2-2025-12-11"

    • "gpt-5.2-chat-latest"

    • "gpt-5.2-pro"

    • "gpt-5.2-pro-2025-12-11"

    • "gpt-5.1"

    • "gpt-5.1-2025-11-13"

    • "gpt-5.1-codex"

    • "gpt-5.1-mini"

    • "gpt-5.1-chat-latest"

    • "gpt-5"

    • "gpt-5-mini"

    • "gpt-5-nano"

    • "gpt-5-2025-08-07"

    • "gpt-5-mini-2025-08-07"

    • "gpt-5-nano-2025-08-07"

    • "gpt-5-chat-latest"

    • "gpt-4.1"

    • "gpt-4.1-mini"

    • "gpt-4.1-nano"

    • "gpt-4.1-2025-04-14"

    • "gpt-4.1-mini-2025-04-14"

    • "gpt-4.1-nano-2025-04-14"

    • "o4-mini"

    • "o4-mini-2025-04-16"

    • "o3"

    • "o3-2025-04-16"

    • "o3-mini"

    • "o3-mini-2025-01-31"

    • "o1"

    • "o1-2024-12-17"

    • "o1-preview"

    • "o1-preview-2024-09-12"

    • "o1-mini"

    • "o1-mini-2024-09-12"

    • "gpt-4o"

    • "gpt-4o-2024-11-20"

    • "gpt-4o-2024-08-06"

    • "gpt-4o-2024-05-13"

    • "gpt-4o-audio-preview"

    • "gpt-4o-audio-preview-2024-10-01"

    • "gpt-4o-audio-preview-2024-12-17"

    • "gpt-4o-audio-preview-2025-06-03"

    • "gpt-4o-mini-audio-preview"

    • "gpt-4o-mini-audio-preview-2024-12-17"

    • "gpt-4o-search-preview"

    • "gpt-4o-mini-search-preview"

    • "gpt-4o-search-preview-2025-03-11"

    • "gpt-4o-mini-search-preview-2025-03-11"

    • "chatgpt-4o-latest"

    • "codex-mini-latest"

    • "gpt-4o-mini"

    • "gpt-4o-mini-2024-07-18"

    • "gpt-4-turbo"

    • "gpt-4-turbo-2024-04-09"

    • "gpt-4-0125-preview"

    • "gpt-4-turbo-preview"

    • "gpt-4-1106-preview"

    • "gpt-4-vision-preview"

    • "gpt-4"

    • "gpt-4-0314"

    • "gpt-4-0613"

    • "gpt-4-32k"

    • "gpt-4-32k-0314"

    • "gpt-4-32k-0613"

    • "gpt-3.5-turbo"

    • "gpt-3.5-turbo-16k"

    • "gpt-3.5-turbo-0301"

    • "gpt-3.5-turbo-0613"

    • "gpt-3.5-turbo-1106"

    • "gpt-3.5-turbo-0125"

    • "gpt-3.5-turbo-16k-0613"

• description: Optional[str]

  The description of the assistant. The maximum length is 512 characters.

• instructions: Optional[str]

  The system instructions that the assistant uses. The maximum length is 256,000 characters.

• 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.

• name: Optional[str]

  The name of the assistant. The maximum length is 256 characters.

• reasoning_effort: Optional[ReasoningEffort]

  Constrains effort on reasoning for reasoning models. Currently supported values are none, minimal, low, medium, high, xhigh, and max. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response. Not all reasoning models support every value. See the reasoning guide for model-specific support.

  • "none"

  • "minimal"

  • "low"

  • "medium"

  • "high"

  • "xhigh"

  • "max"

• response_format: Optional[AssistantResponseFormatOptionParam]

  Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

  Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

  Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

  Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

  • Literal["auto"]

    auto is the default value

    • "auto"

  • class ResponseFormatText: …

    Default response format. Used to generate text responses.

    • type: Literal["text"]

      The type of response format being defined. Always text.

      • "text"

  • class ResponseFormatJSONObject: …

    JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

    • type: Literal["json_object"]

      The type of response format being defined. Always json_object.

      • "json_object"

  • class ResponseFormatJSONSchema: …

    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

    • json_schema: JSONSchema

      Structured Outputs configuration options, including a JSON Schema.

      • name: str

        The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

      • description: Optional[str]

        A description of what the response format is for, used by the model to determine how to respond in the format.

      • schema: Optional[Dict[str, object]]

        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

      • strict: Optional[bool]

        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

    • type: Literal["json_schema"]

      The type of response format being defined. Always json_schema.

      • "json_schema"

• temperature: Optional[float]

  What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

• tool_resources: Optional[ToolResources]

  A set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

  • code_interpreter: Optional[ToolResourcesCodeInterpreter]

    • file_ids: Optional[Sequence[str]]

      A list of file IDs made available to the code_interpreter tool. There can be a maximum of 20 files associated with the tool.

  • file_search: Optional[ToolResourcesFileSearch]

    • vector_store_ids: Optional[Sequence[str]]

      The vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

    • vector_stores: Optional[Iterable[ToolResourcesFileSearchVectorStore]]

      A helper to create a vector store with file_ids and attach it to this assistant. There can be a maximum of 1 vector store attached to the assistant.

      • chunking_strategy: Optional[ToolResourcesFileSearchVectorStoreChunkingStrategy]

        The chunking strategy used to chunk the file(s). If not set, will use the auto strategy.

        • class ToolResourcesFileSearchVectorStoreChunkingStrategyAuto: …

          The default strategy. This strategy currently uses a max_chunk_size_tokens of 800 and chunk_overlap_tokens of 400.

          • type: Literal["auto"]

            Always auto.

            • "auto"

        • class ToolResourcesFileSearchVectorStoreChunkingStrategyStatic: …

          • static: ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic

            • chunk_overlap_tokens: int

              The number of tokens that overlap between chunks. The default value is 400.

              Note that the overlap must not exceed half of max_chunk_size_tokens.

            • max_chunk_size_tokens: int

              The maximum number of tokens in each chunk. The default value is 800. The minimum value is 100 and the maximum value is 4096.

          • type: Literal["static"]

            Always static.

            • "static"

      • file_ids: Optional[Sequence[str]]

        A list of file IDs to add to the vector store. For vector stores created before Nov 2025, there can be a maximum of 10,000 files in a vector store. For vector stores created starting in Nov 2025, the limit is 100,000,000 files.

      • 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.

• tools: Optional[Iterable[object]]

  A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types code_interpreter, file_search, or function.

• top_p: Optional[float]

  An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

  We generally recommend altering this or temperature but not both.

Returns

• class Assistant: …

  Represents an assistant that can call the model and use tools.

  • id: str

    The identifier, which can be referenced in API endpoints.

  • created_at: int

    The Unix timestamp (in seconds) for when the assistant was created.

  • description: Optional[str]

    The description of the assistant. The maximum length is 512 characters.

  • instructions: Optional[str]

    The system instructions that the assistant uses. The maximum length is 256,000 characters.

  • 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.

  • model: str

    ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.

  • name: Optional[str]

    The name of the assistant. The maximum length is 256 characters.

  • object: Literal["assistant"]

    The object type, which is always assistant.

    • "assistant"

  • tools: List[object]

    A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types code_interpreter, file_search, or function.

  • response_format: Optional[AssistantResponseFormatOption]

    Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

    Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

    Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

    Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

    • Literal["auto"]

      auto is the default value

      • "auto"

    • class ResponseFormatText: …

      Default response format. Used to generate text responses.

      • type: Literal["text"]

        The type of response format being defined. Always text.

        • "text"

    • class ResponseFormatJSONObject: …

      JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

      • type: Literal["json_object"]

        The type of response format being defined. Always json_object.

        • "json_object"

    • class ResponseFormatJSONSchema: …

      JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

      • json_schema: JSONSchema

        Structured Outputs configuration options, including a JSON Schema.

        • name: str

          The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

        • description: Optional[str]

          A description of what the response format is for, used by the model to determine how to respond in the format.

        • schema: Optional[Dict[str, object]]

          The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

        • strict: Optional[bool]

          Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

      • type: Literal["json_schema"]

        The type of response format being defined. Always json_schema.

        • "json_schema"

  • temperature: Optional[float]

    What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

  • tool_resources: Optional[ToolResources]

    A set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

    • code_interpreter: Optional[ToolResourcesCodeInterpreter]

      • file_ids: Optional[List[str]]

        A list of file IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

    • file_search: Optional[ToolResourcesFileSearch]

      • vector_store_ids: Optional[List[str]]

        The ID of the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

  • top_p: Optional[float]

    An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

    We generally recommend altering this or temperature but not both.

Example

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ.get("OPENAI_API_KEY"),  # This is the default and can be omitted
)
assistant = client.beta.assistants.create(
    model="gpt-4o",
)
print(assistant.id)

Response

{
  "id": "id",
  "created_at": 0,
  "description": "description",
  "instructions": "instructions",
  "metadata": {
    "foo": "string"
  },
  "model": "model",
  "name": "name",
  "object": "assistant",
  "tools": [
    {}
  ],
  "response_format": "auto",
  "temperature": 1,
  "tool_resources": {
    "code_interpreter": {
      "file_ids": [
        "string"
      ]
    },
    "file_search": {
      "vector_store_ids": [
        "string"
      ]
    }
  },
  "top_p": 1
}

Code Interpreter

from openai import OpenAI
client = OpenAI()

my_assistant = client.beta.assistants.create(
    instructions="You are a personal math tutor. When asked a question, write and run Python code to answer the question.",
    name="Math Tutor",
    tools=[{"type": "code_interpreter"}],
    model="gpt-4o",
)
print(my_assistant)

Response

{
  "id": "asst_abc123",
  "object": "assistant",
  "created_at": 1698984975,
  "name": "Math Tutor",
  "description": null,
  "model": "gpt-4o",
  "instructions": "You are a personal math tutor. When asked a question, write and run Python code to answer the question.",
  "tools": [
    {
      "type": "code_interpreter"
    }
  ],
  "metadata": {},
  "top_p": 1.0,
  "temperature": 1.0,
  "response_format": "auto"
}

Files

from openai import OpenAI
client = OpenAI()

my_assistant = client.beta.assistants.create(
    instructions="You are an HR bot, and you have access to files to answer employee questions about company policies.",
    name="HR Helper",
    tools=[{"type": "file_search"}],
    tool_resources={"file_search": {"vector_store_ids": ["vs_123"]}},
    model="gpt-4o"
)
print(my_assistant)

Response

{
  "id": "asst_abc123",
  "object": "assistant",
  "created_at": 1699009403,
  "name": "HR Helper",
  "description": null,
  "model": "gpt-4o",
  "instructions": "You are an HR bot, and you have access to files to answer employee questions about company policies.",
  "tools": [
    {
      "type": "file_search"
    }
  ],
  "tool_resources": {
    "file_search": {
      "vector_store_ids": ["vs_123"]
    }
  },
  "metadata": {},
  "top_p": 1.0,
  "temperature": 1.0,
  "response_format": "auto"
}

Retrieve assistant

beta.assistants.retrieve(strassistant_id) -> Assistant

get /assistants/{assistant_id}

Retrieve assistant

Parameters

• assistant_id: str

Returns

• class Assistant: …

  Represents an assistant that can call the model and use tools.

  • id: str

    The identifier, which can be referenced in API endpoints.

  • created_at: int

    The Unix timestamp (in seconds) for when the assistant was created.

  • description: Optional[str]

    The description of the assistant. The maximum length is 512 characters.

  • instructions: Optional[str]

    The system instructions that the assistant uses. The maximum length is 256,000 characters.

  • 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.

  • model: str

    ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.

  • name: Optional[str]

    The name of the assistant. The maximum length is 256 characters.

  • object: Literal["assistant"]

    The object type, which is always assistant.

    • "assistant"

  • tools: List[object]

    A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types code_interpreter, file_search, or function.

  • response_format: Optional[AssistantResponseFormatOption]

    Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

    Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

    Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

    Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

    • Literal["auto"]

      auto is the default value

      • "auto"

    • class ResponseFormatText: …

      Default response format. Used to generate text responses.

      • type: Literal["text"]

        The type of response format being defined. Always text.

        • "text"

    • class ResponseFormatJSONObject: …

      JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

      • type: Literal["json_object"]

        The type of response format being defined. Always json_object.

        • "json_object"

    • class ResponseFormatJSONSchema: …

      JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

      • json_schema: JSONSchema

        Structured Outputs configuration options, including a JSON Schema.

        • name: str

          The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

        • description: Optional[str]

          A description of what the response format is for, used by the model to determine how to respond in the format.

        • schema: Optional[Dict[str, object]]

          The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

        • strict: Optional[bool]

          Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

      • type: Literal["json_schema"]

        The type of response format being defined. Always json_schema.

        • "json_schema"

  • temperature: Optional[float]

    What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

  • tool_resources: Optional[ToolResources]

    A set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

    • code_interpreter: Optional[ToolResourcesCodeInterpreter]

      • file_ids: Optional[List[str]]

        A list of file IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

    • file_search: Optional[ToolResourcesFileSearch]

      • vector_store_ids: Optional[List[str]]

        The ID of the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

  • top_p: Optional[float]

    An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

    We generally recommend altering this or temperature but not both.

Example

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ.get("OPENAI_API_KEY"),  # This is the default and can be omitted
)
assistant = client.beta.assistants.retrieve(
    "assistant_id",
)
print(assistant.id)

Response

{
  "id": "id",
  "created_at": 0,
  "description": "description",
  "instructions": "instructions",
  "metadata": {
    "foo": "string"
  },
  "model": "model",
  "name": "name",
  "object": "assistant",
  "tools": [
    {}
  ],
  "response_format": "auto",
  "temperature": 1,
  "tool_resources": {
    "code_interpreter": {
      "file_ids": [
        "string"
      ]
    },
    "file_search": {
      "vector_store_ids": [
        "string"
      ]
    }
  },
  "top_p": 1
}

Example

from openai import OpenAI
client = OpenAI()

my_assistant = client.beta.assistants.retrieve("asst_abc123")
print(my_assistant)

Response

{
  "id": "asst_abc123",
  "object": "assistant",
  "created_at": 1699009709,
  "name": "HR Helper",
  "description": null,
  "model": "gpt-4o",
  "instructions": "You are an HR bot, and you have access to files to answer employee questions about company policies.",
  "tools": [
    {
      "type": "file_search"
    }
  ],
  "metadata": {},
  "top_p": 1.0,
  "temperature": 1.0,
  "response_format": "auto"
}

Modify assistant

beta.assistants.update(strassistant_id, AssistantUpdateParams**kwargs) -> Assistant

post /assistants/{assistant_id}

Modify assistant

Parameters

• assistant_id: str

• description: Optional[str]

  The description of the assistant. The maximum length is 512 characters.

• instructions: Optional[str]

  The system instructions that the assistant uses. The maximum length is 256,000 characters.

• 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.

• model: Optional[Union[str, Literal["gpt-5", "gpt-5-mini", "gpt-5-nano", 39 more]]]

  ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.

  • str

  • Literal["gpt-5", "gpt-5-mini", "gpt-5-nano", 39 more]

    ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.

    • "gpt-5"

    • "gpt-5-mini"

    • "gpt-5-nano"

    • "gpt-5-2025-08-07"

    • "gpt-5-mini-2025-08-07"

    • "gpt-5-nano-2025-08-07"

    • "gpt-4.1"

    • "gpt-4.1-mini"

    • "gpt-4.1-nano"

    • "gpt-4.1-2025-04-14"

    • "gpt-4.1-mini-2025-04-14"

    • "gpt-4.1-nano-2025-04-14"

    • "o3-mini"

    • "o3-mini-2025-01-31"

    • "o1"

    • "o1-2024-12-17"

    • "gpt-4o"

    • "gpt-4o-2024-11-20"

    • "gpt-4o-2024-08-06"

    • "gpt-4o-2024-05-13"

    • "gpt-4o-mini"

    • "gpt-4o-mini-2024-07-18"

    • "gpt-4.5-preview"

    • "gpt-4.5-preview-2025-02-27"

    • "gpt-4-turbo"

    • "gpt-4-turbo-2024-04-09"

    • "gpt-4-0125-preview"

    • "gpt-4-turbo-preview"

    • "gpt-4-1106-preview"

    • "gpt-4-vision-preview"

    • "gpt-4"

    • "gpt-4-0314"

    • "gpt-4-0613"

    • "gpt-4-32k"

    • "gpt-4-32k-0314"

    • "gpt-4-32k-0613"

    • "gpt-3.5-turbo"

    • "gpt-3.5-turbo-16k"

    • "gpt-3.5-turbo-0613"

    • "gpt-3.5-turbo-1106"

    • "gpt-3.5-turbo-0125"

    • "gpt-3.5-turbo-16k-0613"

• name: Optional[str]

  The name of the assistant. The maximum length is 256 characters.

• reasoning_effort: Optional[ReasoningEffort]

  Constrains effort on reasoning for reasoning models. Currently supported values are none, minimal, low, medium, high, xhigh, and max. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response. Not all reasoning models support every value. See the reasoning guide for model-specific support.

  • "none"

  • "minimal"

  • "low"

  • "medium"

  • "high"

  • "xhigh"

  • "max"

• response_format: Optional[AssistantResponseFormatOptionParam]

  Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

  Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

  Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

  Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

  • Literal["auto"]

    auto is the default value

    • "auto"

  • class ResponseFormatText: …

    Default response format. Used to generate text responses.

    • type: Literal["text"]

      The type of response format being defined. Always text.

      • "text"

  • class ResponseFormatJSONObject: …

    JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

    • type: Literal["json_object"]

      The type of response format being defined. Always json_object.

      • "json_object"

  • class ResponseFormatJSONSchema: …

    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

    • json_schema: JSONSchema

      Structured Outputs configuration options, including a JSON Schema.

      • name: str

        The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

      • description: Optional[str]

        A description of what the response format is for, used by the model to determine how to respond in the format.

      • schema: Optional[Dict[str, object]]

        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

      • strict: Optional[bool]

        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

    • type: Literal["json_schema"]

      The type of response format being defined. Always json_schema.

      • "json_schema"

• temperature: Optional[float]

  What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

• tool_resources: Optional[ToolResources]

  A set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

  • code_interpreter: Optional[ToolResourcesCodeInterpreter]

    • file_ids: Optional[Sequence[str]]

      Overrides the list of file IDs made available to the code_interpreter tool. There can be a maximum of 20 files associated with the tool.

  • file_search: Optional[ToolResourcesFileSearch]

    • vector_store_ids: Optional[Sequence[str]]

      Overrides the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

• tools: Optional[Iterable[object]]

  A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types code_interpreter, file_search, or function.

• top_p: Optional[float]

  An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

  We generally recommend altering this or temperature but not both.

Returns

• class Assistant: …

  Represents an assistant that can call the model and use tools.

  • id: str

    The identifier, which can be referenced in API endpoints.

  • created_at: int

    The Unix timestamp (in seconds) for when the assistant was created.

  • description: Optional[str]

    The description of the assistant. The maximum length is 512 characters.

  • instructions: Optional[str]

    The system instructions that the assistant uses. The maximum length is 256,000 characters.

  • 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.

  • model: str

    ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.

  • name: Optional[str]

    The name of the assistant. The maximum length is 256 characters.

  • object: Literal["assistant"]

    The object type, which is always assistant.

    • "assistant"

  • tools: List[object]

    A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types code_interpreter, file_search, or function.

  • response_format: Optional[AssistantResponseFormatOption]

    Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

    Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

    Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

    Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

    • Literal["auto"]

      auto is the default value

      • "auto"

    • class ResponseFormatText: …

      Default response format. Used to generate text responses.

      • type: Literal["text"]

        The type of response format being defined. Always text.

        • "text"

    • class ResponseFormatJSONObject: …

      JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

      • type: Literal["json_object"]

        The type of response format being defined. Always json_object.

        • "json_object"

    • class ResponseFormatJSONSchema: …

      JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

      • json_schema: JSONSchema

        Structured Outputs configuration options, including a JSON Schema.

        • name: str

          The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

        • description: Optional[str]

          A description of what the response format is for, used by the model to determine how to respond in the format.

        • schema: Optional[Dict[str, object]]

          The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

        • strict: Optional[bool]

          Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

      • type: Literal["json_schema"]

        The type of response format being defined. Always json_schema.

        • "json_schema"

  • temperature: Optional[float]

    What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

  • tool_resources: Optional[ToolResources]

    A set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

    • code_interpreter: Optional[ToolResourcesCodeInterpreter]

      • file_ids: Optional[List[str]]

        A list of file IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

    • file_search: Optional[ToolResourcesFileSearch]

      • vector_store_ids: Optional[List[str]]

        The ID of the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

  • top_p: Optional[float]

    An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

    We generally recommend altering this or temperature but not both.

Example

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ.get("OPENAI_API_KEY"),  # This is the default and can be omitted
)
assistant = client.beta.assistants.update(
    assistant_id="assistant_id",
)
print(assistant.id)

Response

{
  "id": "id",
  "created_at": 0,
  "description": "description",
  "instructions": "instructions",
  "metadata": {
    "foo": "string"
  },
  "model": "model",
  "name": "name",
  "object": "assistant",
  "tools": [
    {}
  ],
  "response_format": "auto",
  "temperature": 1,
  "tool_resources": {
    "code_interpreter": {
      "file_ids": [
        "string"
      ]
    },
    "file_search": {
      "vector_store_ids": [
        "string"
      ]
    }
  },
  "top_p": 1
}

Example

from openai import OpenAI
client = OpenAI()

my_updated_assistant = client.beta.assistants.update(
  "asst_abc123",
  instructions="You are an HR bot, and you have access to files to answer employee questions about company policies. Always response with info from either of the files.",
  name="HR Helper",
  tools=[{"type": "file_search"}],
  model="gpt-4o"
)

print(my_updated_assistant)

Response

{
  "id": "asst_123",
  "object": "assistant",
  "created_at": 1699009709,
  "name": "HR Helper",
  "description": null,
  "model": "gpt-4o",
  "instructions": "You are an HR bot, and you have access to files to answer employee questions about company policies. Always response with info from either of the files.",
  "tools": [
    {
      "type": "file_search"
    }
  ],
  "tool_resources": {
    "file_search": {
      "vector_store_ids": []
    }
  },
  "metadata": {},
  "top_p": 1.0,
  "temperature": 1.0,
  "response_format": "auto"
}

Delete assistant

beta.assistants.delete(strassistant_id) -> AssistantDeleted

delete /assistants/{assistant_id}

Delete assistant

Parameters

• assistant_id: str

Returns

• class AssistantDeleted: …

  • id: str

  • deleted: bool

  • object: Literal["assistant.deleted"]

    • "assistant.deleted"

Example

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ.get("OPENAI_API_KEY"),  # This is the default and can be omitted
)
assistant_deleted = client.beta.assistants.delete(
    "assistant_id",
)
print(assistant_deleted.id)

Response

{
  "id": "id",
  "deleted": true,
  "object": "assistant.deleted"
}

Example

from openai import OpenAI
client = OpenAI()

response = client.beta.assistants.delete("asst_abc123")
print(response)

Response

{
  "id": "asst_abc123",
  "object": "assistant.deleted",
  "deleted": true
}

Domain Types

Assistant

• class Assistant: …

  Represents an assistant that can call the model and use tools.

  • id: str

    The identifier, which can be referenced in API endpoints.

  • created_at: int

    The Unix timestamp (in seconds) for when the assistant was created.

  • description: Optional[str]

    The description of the assistant. The maximum length is 512 characters.

  • instructions: Optional[str]

    The system instructions that the assistant uses. The maximum length is 256,000 characters.

  • 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.

  • model: str

    ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.

  • name: Optional[str]

    The name of the assistant. The maximum length is 256 characters.

  • object: Literal["assistant"]

    The object type, which is always assistant.

    • "assistant"

  • tools: List[object]

    A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types code_interpreter, file_search, or function.

  • response_format: Optional[AssistantResponseFormatOption]

    Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

    Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

    Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

    Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

    • Literal["auto"]

      auto is the default value

      • "auto"

    • class ResponseFormatText: …

      Default response format. Used to generate text responses.

      • type: Literal["text"]

        The type of response format being defined. Always text.

        • "text"

    • class ResponseFormatJSONObject: …

      JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

      • type: Literal["json_object"]

        The type of response format being defined. Always json_object.

        • "json_object"

    • class ResponseFormatJSONSchema: …

      JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

      • json_schema: JSONSchema

        Structured Outputs configuration options, including a JSON Schema.

        • name: str

          The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

        • description: Optional[str]

          A description of what the response format is for, used by the model to determine how to respond in the format.

        • schema: Optional[Dict[str, object]]

          The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

        • strict: Optional[bool]

          Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

      • type: Literal["json_schema"]

        The type of response format being defined. Always json_schema.

        • "json_schema"

  • temperature: Optional[float]

    What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

  • tool_resources: Optional[ToolResources]

    A set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

    • code_interpreter: Optional[ToolResourcesCodeInterpreter]

      • file_ids: Optional[List[str]]

        A list of file IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

    • file_search: Optional[ToolResourcesFileSearch]

      • vector_store_ids: Optional[List[str]]

        The ID of the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

  • top_p: Optional[float]

    An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

    We generally recommend altering this or temperature but not both.

Assistant Deleted

• class AssistantDeleted: …

  • id: str

  • deleted: bool

  • object: Literal["assistant.deleted"]

    • "assistant.deleted"

Assistant Stream Event

• AssistantStreamEvent

  Represents an event emitted when streaming a Run.

  Each event in a server-sent events stream has an event and data property:

  event: thread.created
  data: {"id": "thread_123", "object": "thread", ...}
  

  We emit events whenever a new object is created, transitions to a new state, or is being streamed in parts (deltas). For example, we emit thread.run.created when a new run is created, thread.run.completed when a run completes, and so on. When an Assistant chooses to create a message during a run, we emit a thread.message.created event, a thread.message.in_progress event, many thread.message.delta events, and finally a thread.message.completed event.

  We may add additional events over time, so we recommend handling unknown events gracefully in your code. See the Assistants API quickstart to learn how to integrate the Assistants API with streaming.

  • class ThreadCreated: …

    Occurs when a new thread is created.

    • data: Thread

      Represents a thread that contains messages.

      • id: str

        The identifier, which can be referenced in API endpoints.

      • created_at: int

        The Unix timestamp (in seconds) for when the thread was created.

      • 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: Literal["thread"]

        The object type, which is always thread.

        • "thread"

      • tool_resources: Optional[ToolResources]

        A set of resources that are made available to the assistant's tools in this thread. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

        • code_interpreter: Optional[ToolResourcesCodeInterpreter]

          • file_ids: Optional[List[str]]

            A list of file IDs made available to the code_interpreter tool. There can be a maximum of 20 files associated with the tool.

        • file_search: Optional[ToolResourcesFileSearch]

          • vector_store_ids: Optional[List[str]]

            The vector store attached to this thread. There can be a maximum of 1 vector store attached to the thread.

    • event: Literal["thread.created"]

      • "thread.created"

    • enabled: Optional[bool]

      Whether to enable input audio transcription.

  • class ThreadRunCreated: …

    Occurs when a new run is created.

    • data: Run

      Represents an execution run on a thread.

      • id: str

        The identifier, which can be referenced in API endpoints.

      • assistant_id: str

        The ID of the assistant used for execution of this run.

      • cancelled_at: Optional[int]

        The Unix timestamp (in seconds) for when the run was cancelled.

      • completed_at: Optional[int]

        The Unix timestamp (in seconds) for when the run was completed.

      • created_at: int

        The Unix timestamp (in seconds) for when the run was created.

      • expires_at: Optional[int]

        The Unix timestamp (in seconds) for when the run will expire.

      • failed_at: Optional[int]

        The Unix timestamp (in seconds) for when the run failed.

      • incomplete_details: Optional[IncompleteDetails]

        Details on why the run is incomplete. Will be null if the run is not incomplete.

        • reason: Optional[Literal["max_completion_tokens", "max_prompt_tokens"]]

          The reason why the run is incomplete. This will point to which specific token limit was reached over the course of the run.

          • "max_completion_tokens"

          • "max_prompt_tokens"

      • instructions: str

        The instructions that the assistant used for this run.

      • last_error: Optional[LastError]

        The last error associated with this run. Will be null if there are no errors.

        • code: Literal["server_error", "rate_limit_exceeded", "invalid_prompt"]

          One of server_error, rate_limit_exceeded, or invalid_prompt.

          • "server_error"

          • "rate_limit_exceeded"

          • "invalid_prompt"

        • message: str

          A human-readable description of the error.

      • max_completion_tokens: Optional[int]

        The maximum number of completion tokens specified to have been used over the course of the run.

      • max_prompt_tokens: Optional[int]

        The maximum number of prompt tokens specified to have been used over the course of the run.

      • 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.

      • model: str

        The model that the assistant used for this run.

      • object: Literal["thread.run"]

        The object type, which is always thread.run.

        • "thread.run"

      • parallel_tool_calls: bool

        Whether to enable parallel function calling during tool use.

      • required_action: Optional[RequiredAction]

        Details on the action required to continue the run. Will be null if no action is required.

        • submit_tool_outputs: RequiredActionSubmitToolOutputs

          Details on the tool outputs needed for this run to continue.

          • tool_calls: List[RequiredActionFunctionToolCall]

            A list of the relevant tool calls.

            • id: str

              The ID of the tool call. This ID must be referenced when you submit the tool outputs in using the Submit tool outputs to run endpoint.

            • function: Function

              The function definition.

              • arguments: str

                The arguments that the model expects you to pass to the function.

              • name: str

                The name of the function.

            • type: Literal["function"]

              The type of tool call the output is required for. For now, this is always function.

              • "function"

        • type: Literal["submit_tool_outputs"]

          For now, this is always submit_tool_outputs.

          • "submit_tool_outputs"

      • response_format: Optional[AssistantResponseFormatOption]

        Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

        Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

        Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

        Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

        • Literal["auto"]

          auto is the default value

          • "auto"

        • class ResponseFormatText: …

          Default response format. Used to generate text responses.

          • type: Literal["text"]

            The type of response format being defined. Always text.

            • "text"

        • class ResponseFormatJSONObject: …

          JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

          • type: Literal["json_object"]

            The type of response format being defined. Always json_object.

            • "json_object"

        • class ResponseFormatJSONSchema: …

          JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

          • json_schema: JSONSchema

            Structured Outputs configuration options, including a JSON Schema.

            • name: str

              The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

            • description: Optional[str]

              A description of what the response format is for, used by the model to determine how to respond in the format.

            • schema: Optional[Dict[str, object]]

              The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

            • strict: Optional[bool]

              Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

          • type: Literal["json_schema"]

            The type of response format being defined. Always json_schema.

            • "json_schema"

      • started_at: Optional[int]

        The Unix timestamp (in seconds) for when the run was started.

      • status: object

      • thread_id: str

        The ID of the thread that was executed on as a part of this run.

      • tool_choice: Optional[AssistantToolChoiceOption]

        Controls which (if any) tool is called by the model. none means the model will not call any tools and instead generates a message. auto is the default value and means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools before responding to the user. Specifying a particular tool like {"type": "file_search"} or {"type": "function", "function": {"name": "my_function"}} forces the model to call that tool.

        • Literal["none", "auto", "required"]

          none means the model will not call any tools and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools before responding to the user.

          • "none"

          • "auto"

          • "required"

        • class AssistantToolChoice: …

          Specifies a tool the model should use. Use to force the model to call a specific tool.

          • type: Literal["function", "code_interpreter", "file_search"]

            The type of the tool. If type is function, the function name must be set

            • "function"

            • "code_interpreter"

            • "file_search"

          • function: Optional[AssistantToolChoiceFunction]

            • name: str

              The name of the function to call.

      • tools: List[object]

        The list of tools that the assistant used for this run.

      • truncation_strategy: Optional[TruncationStrategy]

        Controls for how a thread will be truncated prior to the run. Use this to control the initial context window of the run.

        • type: Literal["auto", "last_messages"]

          The truncation strategy to use for the thread. The default is auto. If set to last_messages, the thread will be truncated to the n most recent messages in the thread. When set to auto, messages in the middle of the thread will be dropped to fit the context length of the model, max_prompt_tokens.

          • "auto"

          • "last_messages"

        • last_messages: Optional[int]

          The number of most recent messages from the thread when constructing the context for the run.

      • usage: Optional[Usage]

        Usage statistics related to the run. This value will be null if the run is not in a terminal state (i.e. in_progress, queued, etc.).

        • completion_tokens: int

          Number of completion tokens used over the course of the run.

        • prompt_tokens: int

          Number of prompt tokens used over the course of the run.

        • total_tokens: int

          Total number of tokens used (prompt + completion).

      • temperature: Optional[float]

        The sampling temperature used for this run. If not set, defaults to 1.

      • top_p: Optional[float]

        The nucleus sampling value used for this run. If not set, defaults to 1.

    • event: Literal["thread.run.created"]

      • "thread.run.created"

  • class ThreadRunQueued: …

    Occurs when a run moves to a queued status.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.queued"]

      • "thread.run.queued"

  • class ThreadRunInProgress: …

    Occurs when a run moves to an in_progress status.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.in_progress"]

      • "thread.run.in_progress"

  • class ThreadRunRequiresAction: …

    Occurs when a run moves to a requires_action status.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.requires_action"]

      • "thread.run.requires_action"

  • class ThreadRunCompleted: …

    Occurs when a run is completed.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.completed"]

      • "thread.run.completed"

  • class ThreadRunIncomplete: …

    Occurs when a run ends with status incomplete.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.incomplete"]

      • "thread.run.incomplete"

  • class ThreadRunFailed: …

    Occurs when a run fails.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.failed"]

      • "thread.run.failed"

  • class ThreadRunCancelling: …

    Occurs when a run moves to a cancelling status.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.cancelling"]

      • "thread.run.cancelling"

  • class ThreadRunCancelled: …

    Occurs when a run is cancelled.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.cancelled"]

      • "thread.run.cancelled"

  • class ThreadRunExpired: …

    Occurs when a run expires.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.expired"]

      • "thread.run.expired"

  • class ThreadRunStepCreated: …

    Occurs when a run step is created.

    • data: RunStep

      Represents a step in execution of a run.

      • id: str

        The identifier of the run step, which can be referenced in API endpoints.

      • assistant_id: str

        The ID of the assistant associated with the run step.

      • cancelled_at: Optional[int]

        The Unix timestamp (in seconds) for when the run step was cancelled.

      • completed_at: Optional[int]

        The Unix timestamp (in seconds) for when the run step completed.

      • created_at: int

        The Unix timestamp (in seconds) for when the run step was created.

      • expired_at: Optional[int]

        The Unix timestamp (in seconds) for when the run step expired. A step is considered expired if the parent run is expired.

      • failed_at: Optional[int]

        The Unix timestamp (in seconds) for when the run step failed.

      • last_error: Optional[LastError]

        The last error associated with this run step. Will be null if there are no errors.

        • code: Literal["server_error", "rate_limit_exceeded"]

          One of server_error or rate_limit_exceeded.

          • "server_error"

          • "rate_limit_exceeded"

        • message: str

          A human-readable description of the error.

      • 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: Literal["thread.run.step"]

        The object type, which is always thread.run.step.

        • "thread.run.step"

      • run_id: str

        The ID of the run that this run step is a part of.

      • status: Literal["in_progress", "cancelled", "failed", 2 more]

        The status of the run step, which can be either in_progress, cancelled, failed, completed, or expired.

        • "in_progress"

        • "cancelled"

        • "failed"

        • "completed"

        • "expired"

      • step_details: StepDetails

        The details of the run step.

        • class MessageCreationStepDetails: …

          Details of the message creation by the run step.

          • message_creation: MessageCreation

            • message_id: str

              The ID of the message that was created by this run step.

          • type: Literal["message_creation"]

            Always message_creation.

            • "message_creation"

        • class ToolCallsStepDetails: …

          Details of the tool call.

          • tool_calls: List[object]

            An array of tool calls the run step was involved in. These can be associated with one of three types of tools: code_interpreter, file_search, or function.

          • type: Literal["tool_calls"]

            Always tool_calls.

            • "tool_calls"

      • thread_id: str

        The ID of the thread that was run.

      • type: Literal["message_creation", "tool_calls"]

        The type of run step, which can be either message_creation or tool_calls.

        • "message_creation"

        • "tool_calls"

      • usage: Optional[Usage]

        Usage statistics related to the run step. This value will be null while the run step's status is in_progress.

        • completion_tokens: int

          Number of completion tokens used over the course of the run step.

        • prompt_tokens: int

          Number of prompt tokens used over the course of the run step.

        • total_tokens: int

          Total number of tokens used (prompt + completion).

    • event: Literal["thread.run.step.created"]

      • "thread.run.step.created"

  • class ThreadRunStepInProgress: …

    Occurs when a run step moves to an in_progress state.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.in_progress"]

      • "thread.run.step.in_progress"

  • class ThreadRunStepDelta: …

    Occurs when parts of a run step are being streamed.

    • data: RunStepDeltaEvent

      Represents a run step delta i.e. any changed fields on a run step during streaming.

      • id: str

        The identifier of the run step, which can be referenced in API endpoints.

      • delta: object

      • object: Literal["thread.run.step.delta"]

        The object type, which is always thread.run.step.delta.

        • "thread.run.step.delta"

    • event: Literal["thread.run.step.delta"]

      • "thread.run.step.delta"

  • class ThreadRunStepCompleted: …

    Occurs when a run step is completed.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.completed"]

      • "thread.run.step.completed"

  • class ThreadRunStepFailed: …

    Occurs when a run step fails.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.failed"]

      • "thread.run.step.failed"

  • class ThreadRunStepCancelled: …

    Occurs when a run step is cancelled.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.cancelled"]

      • "thread.run.step.cancelled"

  • class ThreadRunStepExpired: …

    Occurs when a run step expires.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.expired"]

      • "thread.run.step.expired"

  • class ThreadMessageCreated: …

    Occurs when a message is created.

    • data: Message

      Represents a message within a thread.

      • id: str

        The identifier, which can be referenced in API endpoints.

      • assistant_id: Optional[str]

        If applicable, the ID of the assistant that authored this message.

      • attachments: Optional[List[Attachment]]

        A list of files attached to the message, and the tools they were added to.

        • file_id: Optional[str]

          The ID of the file to attach to the message.

        • tools: Optional[List[AttachmentTool]]

          The tools to add this file to.

          • class CodeInterpreterTool: …

            • type: Literal["code_interpreter"]

              The type of tool being defined: code_interpreter

              • "code_interpreter"

          • class AttachmentToolAssistantToolsFileSearchTypeOnly: …

            • type: Literal["file_search"]

              The type of tool being defined: file_search

              • "file_search"

      • completed_at: Optional[int]

        The Unix timestamp (in seconds) for when the message was completed.

      • content: List[object]

        The content of the message in array of text and/or images.

      • created_at: int

        The Unix timestamp (in seconds) for when the message was created.

      • incomplete_at: Optional[int]

        The Unix timestamp (in seconds) for when the message was marked as incomplete.

      • incomplete_details: Optional[IncompleteDetails]

        On an incomplete message, details about why the message is incomplete.

        • reason: Literal["content_filter", "max_tokens", "run_cancelled", 2 more]

          The reason the message is incomplete.

          • "content_filter"

          • "max_tokens"

          • "run_cancelled"

          • "run_expired"

          • "run_failed"

      • 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: Literal["thread.message"]

        The object type, which is always thread.message.

        • "thread.message"

      • role: Literal["user", "assistant"]

        The entity that produced the message. One of user or assistant.

        • "user"

        • "assistant"

      • run_id: Optional[str]

        The ID of the run associated with the creation of this message. Value is null when messages are created manually using the create message or create thread endpoints.

      • status: Literal["in_progress", "incomplete", "completed"]

        The status of the message, which can be either in_progress, incomplete, or completed.

        • "in_progress"

        • "incomplete"

        • "completed"

      • thread_id: str

        The thread ID that this message belongs to.

    • event: Literal["thread.message.created"]

      • "thread.message.created"

  • class ThreadMessageInProgress: …

    Occurs when a message moves to an in_progress state.

    • data: Message

      Represents a message within a thread.

    • event: Literal["thread.message.in_progress"]

      • "thread.message.in_progress"

  • class ThreadMessageDelta: …

    Occurs when parts of a Message are being streamed.

    • data: MessageDeltaEvent

      Represents a message delta i.e. any changed fields on a message during streaming.

      • id: str

        The identifier of the message, which can be referenced in API endpoints.

      • delta: MessageDelta

        The delta containing the fields that have changed on the Message.

        • content: Optional[List[object]]

          The content of the message in array of text and/or images.

        • role: Optional[Literal["user", "assistant"]]

          The entity that produced the message. One of user or assistant.

          • "user"

          • "assistant"

      • object: Literal["thread.message.delta"]

        The object type, which is always thread.message.delta.

        • "thread.message.delta"

    • event: Literal["thread.message.delta"]

      • "thread.message.delta"

  • class ThreadMessageCompleted: …

    Occurs when a message is completed.

    • data: Message

      Represents a message within a thread.

    • event: Literal["thread.message.completed"]

      • "thread.message.completed"

  • class ThreadMessageIncomplete: …

    Occurs when a message ends before it is completed.

    • data: Message

      Represents a message within a thread.

    • event: Literal["thread.message.incomplete"]

      • "thread.message.incomplete"

  • class ErrorEvent: …

    Occurs when an error occurs. This can happen due to an internal server error or a timeout.

    • data: ErrorObject

      • code: Optional[str]

      • message: str

      • param: Optional[str]

      • type: str

    • event: Literal["error"]

      • "error"

Assistant Tool

• object

Code Interpreter Tool

• class CodeInterpreterTool: …

  • type: Literal["code_interpreter"]

    The type of tool being defined: code_interpreter

    • "code_interpreter"

File Search Tool

• class FileSearchTool: …

  • type: Literal["file_search"]

    The type of tool being defined: file_search

    • "file_search"

  • file_search: Optional[FileSearch]

    Overrides for the file search tool.

    • max_num_results: Optional[int]

      The maximum number of results the file search tool should output. The default is 20 for gpt-4* models and 5 for gpt-3.5-turbo. This number should be between 1 and 50 inclusive.

      Note that the file search tool may output fewer than max_num_results results. See the file search tool documentation for more information.

    • ranking_options: Optional[FileSearchRankingOptions]

      The ranking options for the file search. If not specified, the file search tool will use the auto ranker and a score_threshold of 0.

      See the file search tool documentation for more information.

      • score_threshold: float

        The score threshold for the file search. All values must be a floating point number between 0 and 1.

      • ranker: Optional[Literal["auto", "default_2024_08_21"]]

        The ranker to use for the file search. If not specified will use the auto ranker.

        • "auto"

        • "default_2024_08_21"

Function Tool

• class FunctionTool: …

  • function: FunctionDefinition

    • name: str

      The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

    • description: Optional[str]

      A description of what the function does, used by the model to choose when and how to call the function.

    • parameters: Optional[FunctionParameters]

      The parameters the functions accepts, described as a JSON Schema object. See the guide for examples, and the JSON Schema reference for documentation about the format.

      Omitting parameters defines a function with an empty parameter list.

    • strict: Optional[bool]

      Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.

  • type: Literal["function"]

    The type of tool being defined: function

    • "function"

Message Stream Event

• MessageStreamEvent

  Occurs when a message is created.

  • class ThreadMessageCreated: …

    Occurs when a message is created.

    • data: Message

      Represents a message within a thread.

      • id: str

        The identifier, which can be referenced in API endpoints.

      • assistant_id: Optional[str]

        If applicable, the ID of the assistant that authored this message.

      • attachments: Optional[List[Attachment]]

        A list of files attached to the message, and the tools they were added to.

        • file_id: Optional[str]

          The ID of the file to attach to the message.

        • tools: Optional[List[AttachmentTool]]

          The tools to add this file to.

          • class CodeInterpreterTool: …

            • type: Literal["code_interpreter"]

              The type of tool being defined: code_interpreter

              • "code_interpreter"

          • class AttachmentToolAssistantToolsFileSearchTypeOnly: …

            • type: Literal["file_search"]

              The type of tool being defined: file_search

              • "file_search"

      • completed_at: Optional[int]

        The Unix timestamp (in seconds) for when the message was completed.

      • content: List[object]

        The content of the message in array of text and/or images.

      • created_at: int

        The Unix timestamp (in seconds) for when the message was created.

      • incomplete_at: Optional[int]

        The Unix timestamp (in seconds) for when the message was marked as incomplete.

      • incomplete_details: Optional[IncompleteDetails]

        On an incomplete message, details about why the message is incomplete.

        • reason: Literal["content_filter", "max_tokens", "run_cancelled", 2 more]

          The reason the message is incomplete.

          • "content_filter"

          • "max_tokens"

          • "run_cancelled"

          • "run_expired"

          • "run_failed"

      • 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: Literal["thread.message"]

        The object type, which is always thread.message.

        • "thread.message"

      • role: Literal["user", "assistant"]

        The entity that produced the message. One of user or assistant.

        • "user"

        • "assistant"

      • run_id: Optional[str]

        The ID of the run associated with the creation of this message. Value is null when messages are created manually using the create message or create thread endpoints.

      • status: Literal["in_progress", "incomplete", "completed"]

        The status of the message, which can be either in_progress, incomplete, or completed.

        • "in_progress"

        • "incomplete"

        • "completed"

      • thread_id: str

        The thread ID that this message belongs to.

    • event: Literal["thread.message.created"]

      • "thread.message.created"

  • class ThreadMessageInProgress: …

    Occurs when a message moves to an in_progress state.

    • data: Message

      Represents a message within a thread.

    • event: Literal["thread.message.in_progress"]

      • "thread.message.in_progress"

  • class ThreadMessageDelta: …

    Occurs when parts of a Message are being streamed.

    • data: MessageDeltaEvent

      Represents a message delta i.e. any changed fields on a message during streaming.

      • id: str

        The identifier of the message, which can be referenced in API endpoints.

      • delta: MessageDelta

        The delta containing the fields that have changed on the Message.

        • content: Optional[List[object]]

          The content of the message in array of text and/or images.

        • role: Optional[Literal["user", "assistant"]]

          The entity that produced the message. One of user or assistant.

          • "user"

          • "assistant"

      • object: Literal["thread.message.delta"]

        The object type, which is always thread.message.delta.

        • "thread.message.delta"

    • event: Literal["thread.message.delta"]

      • "thread.message.delta"

  • class ThreadMessageCompleted: …

    Occurs when a message is completed.

    • data: Message

      Represents a message within a thread.

    • event: Literal["thread.message.completed"]

      • "thread.message.completed"

  • class ThreadMessageIncomplete: …

    Occurs when a message ends before it is completed.

    • data: Message

      Represents a message within a thread.

    • event: Literal["thread.message.incomplete"]

      • "thread.message.incomplete"

Run Step Stream Event

• RunStepStreamEvent

  Occurs when a run step is created.

  • class ThreadRunStepCreated: …

    Occurs when a run step is created.

    • data: RunStep

      Represents a step in execution of a run.

      • id: str

        The identifier of the run step, which can be referenced in API endpoints.

      • assistant_id: str

        The ID of the assistant associated with the run step.

      • cancelled_at: Optional[int]

        The Unix timestamp (in seconds) for when the run step was cancelled.

      • completed_at: Optional[int]

        The Unix timestamp (in seconds) for when the run step completed.

      • created_at: int

        The Unix timestamp (in seconds) for when the run step was created.

      • expired_at: Optional[int]

        The Unix timestamp (in seconds) for when the run step expired. A step is considered expired if the parent run is expired.

      • failed_at: Optional[int]

        The Unix timestamp (in seconds) for when the run step failed.

      • last_error: Optional[LastError]

        The last error associated with this run step. Will be null if there are no errors.

        • code: Literal["server_error", "rate_limit_exceeded"]

          One of server_error or rate_limit_exceeded.

          • "server_error"

          • "rate_limit_exceeded"

        • message: str

          A human-readable description of the error.

      • 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: Literal["thread.run.step"]

        The object type, which is always thread.run.step.

        • "thread.run.step"

      • run_id: str

        The ID of the run that this run step is a part of.

      • status: Literal["in_progress", "cancelled", "failed", 2 more]

        The status of the run step, which can be either in_progress, cancelled, failed, completed, or expired.

        • "in_progress"

        • "cancelled"

        • "failed"

        • "completed"

        • "expired"

      • step_details: StepDetails

        The details of the run step.

        • class MessageCreationStepDetails: …

          Details of the message creation by the run step.

          • message_creation: MessageCreation

            • message_id: str

              The ID of the message that was created by this run step.

          • type: Literal["message_creation"]

            Always message_creation.

            • "message_creation"

        • class ToolCallsStepDetails: …

          Details of the tool call.

          • tool_calls: List[object]

            An array of tool calls the run step was involved in. These can be associated with one of three types of tools: code_interpreter, file_search, or function.

          • type: Literal["tool_calls"]

            Always tool_calls.

            • "tool_calls"

      • thread_id: str

        The ID of the thread that was run.

      • type: Literal["message_creation", "tool_calls"]

        The type of run step, which can be either message_creation or tool_calls.

        • "message_creation"

        • "tool_calls"

      • usage: Optional[Usage]

        Usage statistics related to the run step. This value will be null while the run step's status is in_progress.

        • completion_tokens: int

          Number of completion tokens used over the course of the run step.

        • prompt_tokens: int

          Number of prompt tokens used over the course of the run step.

        • total_tokens: int

          Total number of tokens used (prompt + completion).

    • event: Literal["thread.run.step.created"]

      • "thread.run.step.created"

  • class ThreadRunStepInProgress: …

    Occurs when a run step moves to an in_progress state.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.in_progress"]

      • "thread.run.step.in_progress"

  • class ThreadRunStepDelta: …

    Occurs when parts of a run step are being streamed.

    • data: RunStepDeltaEvent

      Represents a run step delta i.e. any changed fields on a run step during streaming.

      • id: str

        The identifier of the run step, which can be referenced in API endpoints.

      • delta: object

      • object: Literal["thread.run.step.delta"]

        The object type, which is always thread.run.step.delta.

        • "thread.run.step.delta"

    • event: Literal["thread.run.step.delta"]

      • "thread.run.step.delta"

  • class ThreadRunStepCompleted: …

    Occurs when a run step is completed.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.completed"]

      • "thread.run.step.completed"

  • class ThreadRunStepFailed: …

    Occurs when a run step fails.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.failed"]

      • "thread.run.step.failed"

  • class ThreadRunStepCancelled: …

    Occurs when a run step is cancelled.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.cancelled"]

      • "thread.run.step.cancelled"

  • class ThreadRunStepExpired: …

    Occurs when a run step expires.

    • data: RunStep

      Represents a step in execution of a run.

    • event: Literal["thread.run.step.expired"]

      • "thread.run.step.expired"

Run Stream Event

• RunStreamEvent

  Occurs when a new run is created.

  • class ThreadRunCreated: …

    Occurs when a new run is created.

    • data: Run

      Represents an execution run on a thread.

      • id: str

        The identifier, which can be referenced in API endpoints.

      • assistant_id: str

        The ID of the assistant used for execution of this run.

      • cancelled_at: Optional[int]

        The Unix timestamp (in seconds) for when the run was cancelled.

      • completed_at: Optional[int]

        The Unix timestamp (in seconds) for when the run was completed.

      • created_at: int

        The Unix timestamp (in seconds) for when the run was created.

      • expires_at: Optional[int]

        The Unix timestamp (in seconds) for when the run will expire.

      • failed_at: Optional[int]

        The Unix timestamp (in seconds) for when the run failed.

      • incomplete_details: Optional[IncompleteDetails]

        Details on why the run is incomplete. Will be null if the run is not incomplete.

        • reason: Optional[Literal["max_completion_tokens", "max_prompt_tokens"]]

          The reason why the run is incomplete. This will point to which specific token limit was reached over the course of the run.

          • "max_completion_tokens"

          • "max_prompt_tokens"

      • instructions: str

        The instructions that the assistant used for this run.

      • last_error: Optional[LastError]

        The last error associated with this run. Will be null if there are no errors.

        • code: Literal["server_error", "rate_limit_exceeded", "invalid_prompt"]

          One of server_error, rate_limit_exceeded, or invalid_prompt.

          • "server_error"

          • "rate_limit_exceeded"

          • "invalid_prompt"

        • message: str

          A human-readable description of the error.

      • max_completion_tokens: Optional[int]

        The maximum number of completion tokens specified to have been used over the course of the run.

      • max_prompt_tokens: Optional[int]

        The maximum number of prompt tokens specified to have been used over the course of the run.

      • 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.

      • model: str

        The model that the assistant used for this run.

      • object: Literal["thread.run"]

        The object type, which is always thread.run.

        • "thread.run"

      • parallel_tool_calls: bool

        Whether to enable parallel function calling during tool use.

      • required_action: Optional[RequiredAction]

        Details on the action required to continue the run. Will be null if no action is required.

        • submit_tool_outputs: RequiredActionSubmitToolOutputs

          Details on the tool outputs needed for this run to continue.

          • tool_calls: List[RequiredActionFunctionToolCall]

            A list of the relevant tool calls.

            • id: str

              The ID of the tool call. This ID must be referenced when you submit the tool outputs in using the Submit tool outputs to run endpoint.

            • function: Function

              The function definition.

              • arguments: str

                The arguments that the model expects you to pass to the function.

              • name: str

                The name of the function.

            • type: Literal["function"]

              The type of tool call the output is required for. For now, this is always function.

              • "function"

        • type: Literal["submit_tool_outputs"]

          For now, this is always submit_tool_outputs.

          • "submit_tool_outputs"

      • response_format: Optional[AssistantResponseFormatOption]

        Specifies the format that the model must output. Compatible with GPT-4o, GPT-4 Turbo, and all GPT-3.5 Turbo models since gpt-3.5-turbo-1106.

        Setting to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the Structured Outputs guide.

        Setting to { "type": "json_object" } enables JSON mode, which ensures the message the model generates is valid JSON.

        Important: when using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if finish_reason="length", which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.

        • Literal["auto"]

          auto is the default value

          • "auto"

        • class ResponseFormatText: …

          Default response format. Used to generate text responses.

          • type: Literal["text"]

            The type of response format being defined. Always text.

            • "text"

        • class ResponseFormatJSONObject: …

          JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.

          • type: Literal["json_object"]

            The type of response format being defined. Always json_object.

            • "json_object"

        • class ResponseFormatJSONSchema: …

          JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.

          • json_schema: JSONSchema

            Structured Outputs configuration options, including a JSON Schema.

            • name: str

              The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

            • description: Optional[str]

              A description of what the response format is for, used by the model to determine how to respond in the format.

            • schema: Optional[Dict[str, object]]

              The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.

            • strict: Optional[bool]

              Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.

          • type: Literal["json_schema"]

            The type of response format being defined. Always json_schema.

            • "json_schema"

      • started_at: Optional[int]

        The Unix timestamp (in seconds) for when the run was started.

      • status: object

      • thread_id: str

        The ID of the thread that was executed on as a part of this run.

      • tool_choice: Optional[AssistantToolChoiceOption]

        Controls which (if any) tool is called by the model. none means the model will not call any tools and instead generates a message. auto is the default value and means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools before responding to the user. Specifying a particular tool like {"type": "file_search"} or {"type": "function", "function": {"name": "my_function"}} forces the model to call that tool.

        • Literal["none", "auto", "required"]

          none means the model will not call any tools and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools before responding to the user.

          • "none"

          • "auto"

          • "required"

        • class AssistantToolChoice: …

          Specifies a tool the model should use. Use to force the model to call a specific tool.

          • type: Literal["function", "code_interpreter", "file_search"]

            The type of the tool. If type is function, the function name must be set

            • "function"

            • "code_interpreter"

            • "file_search"

          • function: Optional[AssistantToolChoiceFunction]

            • name: str

              The name of the function to call.

      • tools: List[object]

        The list of tools that the assistant used for this run.

      • truncation_strategy: Optional[TruncationStrategy]

        Controls for how a thread will be truncated prior to the run. Use this to control the initial context window of the run.

        • type: Literal["auto", "last_messages"]

          The truncation strategy to use for the thread. The default is auto. If set to last_messages, the thread will be truncated to the n most recent messages in the thread. When set to auto, messages in the middle of the thread will be dropped to fit the context length of the model, max_prompt_tokens.

          • "auto"

          • "last_messages"

        • last_messages: Optional[int]

          The number of most recent messages from the thread when constructing the context for the run.

      • usage: Optional[Usage]

        Usage statistics related to the run. This value will be null if the run is not in a terminal state (i.e. in_progress, queued, etc.).

        • completion_tokens: int

          Number of completion tokens used over the course of the run.

        • prompt_tokens: int

          Number of prompt tokens used over the course of the run.

        • total_tokens: int

          Total number of tokens used (prompt + completion).

      • temperature: Optional[float]

        The sampling temperature used for this run. If not set, defaults to 1.

      • top_p: Optional[float]

        The nucleus sampling value used for this run. If not set, defaults to 1.

    • event: Literal["thread.run.created"]

      • "thread.run.created"

  • class ThreadRunQueued: …

    Occurs when a run moves to a queued status.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.queued"]

      • "thread.run.queued"

  • class ThreadRunInProgress: …

    Occurs when a run moves to an in_progress status.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.in_progress"]

      • "thread.run.in_progress"

  • class ThreadRunRequiresAction: …

    Occurs when a run moves to a requires_action status.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.requires_action"]

      • "thread.run.requires_action"

  • class ThreadRunCompleted: …

    Occurs when a run is completed.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.completed"]

      • "thread.run.completed"

  • class ThreadRunIncomplete: …

    Occurs when a run ends with status incomplete.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.incomplete"]

      • "thread.run.incomplete"

  • class ThreadRunFailed: …

    Occurs when a run fails.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.failed"]

      • "thread.run.failed"

  • class ThreadRunCancelling: …

    Occurs when a run moves to a cancelling status.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.cancelling"]

      • "thread.run.cancelling"

  • class ThreadRunCancelled: …

    Occurs when a run is cancelled.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.cancelled"]

      • "thread.run.cancelled"

  • class ThreadRunExpired: …

    Occurs when a run expires.

    • data: Run

      Represents an execution run on a thread.

    • event: Literal["thread.run.expired"]

      • "thread.run.expired"

Thread Stream Event

• class ThreadStreamEvent: …

  Occurs when a new thread is created.

  • data: Thread

    Represents a thread that contains messages.

    • id: str

      The identifier, which can be referenced in API endpoints.

    • created_at: int

      The Unix timestamp (in seconds) for when the thread was created.

    • 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: Literal["thread"]

      The object type, which is always thread.

      • "thread"

    • tool_resources: Optional[ToolResources]

      A set of resources that are made available to the assistant's tools in this thread. The resources are specific to the type of tool. For example, the code_interpreter tool requires a list of file IDs, while the file_search tool requires a list of vector store IDs.

      • code_interpreter: Optional[ToolResourcesCodeInterpreter]

        • file_ids: Optional[List[str]]

          A list of file IDs made available to the code_interpreter tool. There can be a maximum of 20 files associated with the tool.

      • file_search: Optional[ToolResourcesFileSearch]

        • vector_store_ids: Optional[List[str]]

          The vector store attached to this thread. There can be a maximum of 1 vector store attached to the thread.

  • event: Literal["thread.created"]

    • "thread.created"

  • enabled: Optional[bool]

    Whether to enable input audio transcription.