Assistants

List assistants

get /assistants

Returns a list of assistants.

Query Parameters

after: optional string

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 string

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 number

A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.
order: optional "asc" or "desc"

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

Returns

data: array of Assistant
- id: string
  
  The identifier, which can be referenced in API endpoints.
- created_at: number
  
  The Unix timestamp (in seconds) for when the assistant was created.
- description: string
  
  The description of the assistant. The maximum length is 512 characters.
- instructions: string
  
  The system instructions that the assistant uses. The maximum length is 256,000 characters.
- metadata: 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: string
  
  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: string
  
  The name of the assistant. The maximum length is 256 characters.
- object: "assistant"
  
  The object type, which is always assistant.
  - "assistant"
- tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool
  
  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.
  - CodeInterpreterTool object { type }
    - type: "code_interpreter"
      
      The type of tool being defined: code_interpreter
      - "code_interpreter"
  - FileSearchTool object { type, file_search }
    - type: "file_search"
      
      The type of tool being defined: file_search
      - "file_search"
    - file_search: optional object { max_num_results, ranking_options }
      
      Overrides for the file search tool.
      - max_num_results: optional number
        
        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 object { score_threshold, ranker }
        
        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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: optional "auto" or "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"
  - FunctionTool object { function, type }
    - function: FunctionDefinition
      - name: string
        
        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 string
        
        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 boolean
        
        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: "function"
      
      The type of tool being defined: function
      - "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.
  - "auto"
    
    auto is the default value
    - "auto"
  - ResponseFormatText object { type }
    
    Default response format. Used to generate text responses.
    - type: "text"
      
      The type of response format being defined. Always text.
      - "text"
  - ResponseFormatJSONObject object { type }
    
    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: "json_object"
      
      The type of response format being defined. Always json_object.
      - "json_object"
  - ResponseFormatJSONSchema object { json_schema, type }
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: object { name, description, schema, strict }
      
      Structured Outputs configuration options, including a JSON Schema.
      - name: string
        
        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 string
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: optional map[unknown]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: optional boolean
        
        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: "json_schema"
      
      The type of response format being defined. Always json_schema.
      - "json_schema"
- temperature: optional number
  
  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 object { code_interpreter, file_search }
  
  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 object { file_ids }
    - file_ids: optional array of string
      
      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 object { vector_store_ids }
    - vector_store_ids: optional array of string
      
      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 number
  
  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.
first_id: string
has_more: boolean
last_id: string
object: string

Example

curl https://api.openai.com/v1/assistants \
    -H 'OpenAI-Beta: assistants=v2' \
    -H "Authorization: Bearer $OPENAI_API_KEY"

Response

{
  "data": [
    {
      "id": "id",
      "created_at": 0,
      "description": "description",
      "instructions": "instructions",
      "metadata": {
        "foo": "string"
      },
      "model": "model",
      "name": "name",
      "object": "assistant",
      "tools": [
        {
          "type": "code_interpreter"
        }
      ],
      "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

curl "https://api.openai.com/v1/assistants?order=desc&limit=20" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "OpenAI-Beta: assistants=v2"

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

post /assistants

Create an assistant with a model and instructions.

Body Parameters

model: string or "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 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.
- string
- AssistantSupportedModels = "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 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"
description: optional string

The description of the assistant. The maximum length is 512 characters.
instructions: optional string

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 string

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 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.
- "auto"
  
  auto is the default value
  - "auto"
- ResponseFormatText object { type }
  
  Default response format. Used to generate text responses.
  - type: "text"
    
    The type of response format being defined. Always text.
    - "text"
- ResponseFormatJSONObject object { type }
  
  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: "json_object"
    
    The type of response format being defined. Always json_object.
    - "json_object"
- ResponseFormatJSONSchema object { json_schema, type }
  
  JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
  - json_schema: object { name, description, schema, strict }
    
    Structured Outputs configuration options, including a JSON Schema.
    - name: string
      
      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 string
      
      A description of what the response format is for, used by the model to determine how to respond in the format.
    - schema: optional map[unknown]
      
      The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
    - strict: optional boolean
      
      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: "json_schema"
    
    The type of response format being defined. Always json_schema.
    - "json_schema"
temperature: optional number

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 object { code_interpreter, file_search }

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 object { file_ids }
  - file_ids: optional array of string
    
    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 object { vector_store_ids, vector_stores }
  - vector_store_ids: optional array of string
    
    The vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.
  - vector_stores: optional array of object { chunking_strategy, file_ids, metadata }
    
    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 object { type } or object { static, type }
      
      The chunking strategy used to chunk the file(s). If not set, will use the auto strategy.
      - AutoChunkingStrategy object { type }
        
        The default strategy. This strategy currently uses a max_chunk_size_tokens of 800 and chunk_overlap_tokens of 400.
        
        type: "auto"
        
        Always auto.
        
        "auto"
      - StaticChunkingStrategy object { static, type }
        
        static: object { chunk_overlap_tokens, max_chunk_size_tokens }
        
        chunk_overlap_tokens: number
        
        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: number
        
        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: "static"
        
        Always static.
        
        "static"
    - file_ids: optional array of string
      
      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 array of CodeInterpreterTool or FileSearchTool or FunctionTool

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.
- CodeInterpreterTool object { type }
  - type: "code_interpreter"
    
    The type of tool being defined: code_interpreter
    - "code_interpreter"
- FileSearchTool object { type, file_search }
  - type: "file_search"
    
    The type of tool being defined: file_search
    - "file_search"
  - file_search: optional object { max_num_results, ranking_options }
    
    Overrides for the file search tool.
    - max_num_results: optional number
      
      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 object { score_threshold, ranker }
      
      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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
      - ranker: optional "auto" or "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"
- FunctionTool object { function, type }
  - function: FunctionDefinition
    - name: string
      
      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 string
      
      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 boolean
      
      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: "function"
    
    The type of tool being defined: function
    - "function"
top_p: optional number

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

Assistant object { id, created_at, description, 10 more }

Represents an assistant that can call the model and use tools.
- id: string
  
  The identifier, which can be referenced in API endpoints.
- created_at: number
  
  The Unix timestamp (in seconds) for when the assistant was created.
- description: string
  
  The description of the assistant. The maximum length is 512 characters.
- instructions: string
  
  The system instructions that the assistant uses. The maximum length is 256,000 characters.
- metadata: 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: string
  
  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: string
  
  The name of the assistant. The maximum length is 256 characters.
- object: "assistant"
  
  The object type, which is always assistant.
  - "assistant"
- tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool
  
  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.
  - CodeInterpreterTool object { type }
    - type: "code_interpreter"
      
      The type of tool being defined: code_interpreter
      - "code_interpreter"
  - FileSearchTool object { type, file_search }
    - type: "file_search"
      
      The type of tool being defined: file_search
      - "file_search"
    - file_search: optional object { max_num_results, ranking_options }
      
      Overrides for the file search tool.
      - max_num_results: optional number
        
        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 object { score_threshold, ranker }
        
        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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: optional "auto" or "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"
  - FunctionTool object { function, type }
    - function: FunctionDefinition
      - name: string
        
        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 string
        
        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 boolean
        
        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: "function"
      
      The type of tool being defined: function
      - "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.
  - "auto"
    
    auto is the default value
    - "auto"
  - ResponseFormatText object { type }
    
    Default response format. Used to generate text responses.
    - type: "text"
      
      The type of response format being defined. Always text.
      - "text"
  - ResponseFormatJSONObject object { type }
    
    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: "json_object"
      
      The type of response format being defined. Always json_object.
      - "json_object"
  - ResponseFormatJSONSchema object { json_schema, type }
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: object { name, description, schema, strict }
      
      Structured Outputs configuration options, including a JSON Schema.
      - name: string
        
        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 string
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: optional map[unknown]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: optional boolean
        
        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: "json_schema"
      
      The type of response format being defined. Always json_schema.
      - "json_schema"
- temperature: optional number
  
  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 object { code_interpreter, file_search }
  
  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 object { file_ids }
    - file_ids: optional array of string
      
      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 object { vector_store_ids }
    - vector_store_ids: optional array of string
      
      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 number
  
  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

curl https://api.openai.com/v1/assistants \
    -H 'Content-Type: application/json' \
    -H 'OpenAI-Beta: assistants=v2' \
    -H "Authorization: Bearer $OPENAI_API_KEY" \
    -d '{
          "model": "gpt-4o",
          "temperature": 1,
          "top_p": 1
        }'

Response

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

Code Interpreter

curl "https://api.openai.com/v1/assistants" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "OpenAI-Beta: assistants=v2" \
  -d '{
    "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"
  }'

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

curl https://api.openai.com/v1/assistants \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "OpenAI-Beta: assistants=v2" \
  -d '{
    "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"]}},
    "model": "gpt-4o"
  }'

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

get /assistants/{assistant_id}

Retrieves an assistant.

Path Parameters

assistant_id: string

Returns

Assistant object { id, created_at, description, 10 more }

Represents an assistant that can call the model and use tools.
- id: string
  
  The identifier, which can be referenced in API endpoints.
- created_at: number
  
  The Unix timestamp (in seconds) for when the assistant was created.
- description: string
  
  The description of the assistant. The maximum length is 512 characters.
- instructions: string
  
  The system instructions that the assistant uses. The maximum length is 256,000 characters.
- metadata: 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: string
  
  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: string
  
  The name of the assistant. The maximum length is 256 characters.
- object: "assistant"
  
  The object type, which is always assistant.
  - "assistant"
- tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool
  
  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.
  - CodeInterpreterTool object { type }
    - type: "code_interpreter"
      
      The type of tool being defined: code_interpreter
      - "code_interpreter"
  - FileSearchTool object { type, file_search }
    - type: "file_search"
      
      The type of tool being defined: file_search
      - "file_search"
    - file_search: optional object { max_num_results, ranking_options }
      
      Overrides for the file search tool.
      - max_num_results: optional number
        
        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 object { score_threshold, ranker }
        
        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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: optional "auto" or "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"
  - FunctionTool object { function, type }
    - function: FunctionDefinition
      - name: string
        
        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 string
        
        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 boolean
        
        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: "function"
      
      The type of tool being defined: function
      - "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.
  - "auto"
    
    auto is the default value
    - "auto"
  - ResponseFormatText object { type }
    
    Default response format. Used to generate text responses.
    - type: "text"
      
      The type of response format being defined. Always text.
      - "text"
  - ResponseFormatJSONObject object { type }
    
    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: "json_object"
      
      The type of response format being defined. Always json_object.
      - "json_object"
  - ResponseFormatJSONSchema object { json_schema, type }
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: object { name, description, schema, strict }
      
      Structured Outputs configuration options, including a JSON Schema.
      - name: string
        
        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 string
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: optional map[unknown]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: optional boolean
        
        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: "json_schema"
      
      The type of response format being defined. Always json_schema.
      - "json_schema"
- temperature: optional number
  
  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 object { code_interpreter, file_search }
  
  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 object { file_ids }
    - file_ids: optional array of string
      
      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 object { vector_store_ids }
    - vector_store_ids: optional array of string
      
      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 number
  
  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

curl https://api.openai.com/v1/assistants/$ASSISTANT_ID \
    -H 'OpenAI-Beta: assistants=v2' \
    -H "Authorization: Bearer $OPENAI_API_KEY"

Response

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

Example

curl https://api.openai.com/v1/assistants/asst_abc123 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "OpenAI-Beta: assistants=v2"

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

post /assistants/{assistant_id}

Modifies an assistant.

Path Parameters

assistant_id: string

Body Parameters

description: optional string

The description of the assistant. The maximum length is 512 characters.
instructions: optional string

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 string or "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 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.
- string
- AssistantSupportedModels = "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 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 string

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 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.
- "auto"
  
  auto is the default value
  - "auto"
- ResponseFormatText object { type }
  
  Default response format. Used to generate text responses.
  - type: "text"
    
    The type of response format being defined. Always text.
    - "text"
- ResponseFormatJSONObject object { type }
  
  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: "json_object"
    
    The type of response format being defined. Always json_object.
    - "json_object"
- ResponseFormatJSONSchema object { json_schema, type }
  
  JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
  - json_schema: object { name, description, schema, strict }
    
    Structured Outputs configuration options, including a JSON Schema.
    - name: string
      
      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 string
      
      A description of what the response format is for, used by the model to determine how to respond in the format.
    - schema: optional map[unknown]
      
      The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
    - strict: optional boolean
      
      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: "json_schema"
    
    The type of response format being defined. Always json_schema.
    - "json_schema"
temperature: optional number

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 object { code_interpreter, file_search }

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 object { file_ids }
  - file_ids: optional array of string
    
    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 object { vector_store_ids }
  - vector_store_ids: optional array of string
    
    Overrides the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.
tools: optional array of CodeInterpreterTool or FileSearchTool or FunctionTool

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.
- CodeInterpreterTool object { type }
  - type: "code_interpreter"
    
    The type of tool being defined: code_interpreter
    - "code_interpreter"
- FileSearchTool object { type, file_search }
  - type: "file_search"
    
    The type of tool being defined: file_search
    - "file_search"
  - file_search: optional object { max_num_results, ranking_options }
    
    Overrides for the file search tool.
    - max_num_results: optional number
      
      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 object { score_threshold, ranker }
      
      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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
      - ranker: optional "auto" or "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"
- FunctionTool object { function, type }
  - function: FunctionDefinition
    - name: string
      
      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 string
      
      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 boolean
      
      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: "function"
    
    The type of tool being defined: function
    - "function"
top_p: optional number

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

Assistant object { id, created_at, description, 10 more }

Represents an assistant that can call the model and use tools.
- id: string
  
  The identifier, which can be referenced in API endpoints.
- created_at: number
  
  The Unix timestamp (in seconds) for when the assistant was created.
- description: string
  
  The description of the assistant. The maximum length is 512 characters.
- instructions: string
  
  The system instructions that the assistant uses. The maximum length is 256,000 characters.
- metadata: 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: string
  
  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: string
  
  The name of the assistant. The maximum length is 256 characters.
- object: "assistant"
  
  The object type, which is always assistant.
  - "assistant"
- tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool
  
  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.
  - CodeInterpreterTool object { type }
    - type: "code_interpreter"
      
      The type of tool being defined: code_interpreter
      - "code_interpreter"
  - FileSearchTool object { type, file_search }
    - type: "file_search"
      
      The type of tool being defined: file_search
      - "file_search"
    - file_search: optional object { max_num_results, ranking_options }
      
      Overrides for the file search tool.
      - max_num_results: optional number
        
        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 object { score_threshold, ranker }
        
        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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: optional "auto" or "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"
  - FunctionTool object { function, type }
    - function: FunctionDefinition
      - name: string
        
        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 string
        
        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 boolean
        
        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: "function"
      
      The type of tool being defined: function
      - "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.
  - "auto"
    
    auto is the default value
    - "auto"
  - ResponseFormatText object { type }
    
    Default response format. Used to generate text responses.
    - type: "text"
      
      The type of response format being defined. Always text.
      - "text"
  - ResponseFormatJSONObject object { type }
    
    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: "json_object"
      
      The type of response format being defined. Always json_object.
      - "json_object"
  - ResponseFormatJSONSchema object { json_schema, type }
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: object { name, description, schema, strict }
      
      Structured Outputs configuration options, including a JSON Schema.
      - name: string
        
        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 string
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: optional map[unknown]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: optional boolean
        
        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: "json_schema"
      
      The type of response format being defined. Always json_schema.
      - "json_schema"
- temperature: optional number
  
  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 object { code_interpreter, file_search }
  
  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 object { file_ids }
    - file_ids: optional array of string
      
      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 object { vector_store_ids }
    - vector_store_ids: optional array of string
      
      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 number
  
  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

curl https://api.openai.com/v1/assistants/$ASSISTANT_ID \
    -H 'Content-Type: application/json' \
    -H 'OpenAI-Beta: assistants=v2' \
    -H "Authorization: Bearer $OPENAI_API_KEY" \
    -d '{
          "temperature": 1,
          "top_p": 1
        }'

Response

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

Example

curl https://api.openai.com/v1/assistants/asst_abc123 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "OpenAI-Beta: assistants=v2" \
  -d '{
      "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"}],
      "model": "gpt-4o"
    }'

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

delete /assistants/{assistant_id}

Delete an assistant.

Path Parameters

assistant_id: string

Returns

AssistantDeleted object { id, deleted, object }
- id: string
- deleted: boolean
- object: "assistant.deleted"
  - "assistant.deleted"

Example

curl https://api.openai.com/v1/assistants/$ASSISTANT_ID \
    -X DELETE \
    -H 'OpenAI-Beta: assistants=v2' \
    -H "Authorization: Bearer $OPENAI_API_KEY"

Response

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

Example

curl https://api.openai.com/v1/assistants/asst_abc123 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "OpenAI-Beta: assistants=v2" \
  -X DELETE

Response

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

Domain Types

Assistant

Assistant object { id, created_at, description, 10 more }

Represents an assistant that can call the model and use tools.
- id: string
  
  The identifier, which can be referenced in API endpoints.
- created_at: number
  
  The Unix timestamp (in seconds) for when the assistant was created.
- description: string
  
  The description of the assistant. The maximum length is 512 characters.
- instructions: string
  
  The system instructions that the assistant uses. The maximum length is 256,000 characters.
- metadata: 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: string
  
  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: string
  
  The name of the assistant. The maximum length is 256 characters.
- object: "assistant"
  
  The object type, which is always assistant.
  - "assistant"
- tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool
  
  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.
  - CodeInterpreterTool object { type }
    - type: "code_interpreter"
      
      The type of tool being defined: code_interpreter
      - "code_interpreter"
  - FileSearchTool object { type, file_search }
    - type: "file_search"
      
      The type of tool being defined: file_search
      - "file_search"
    - file_search: optional object { max_num_results, ranking_options }
      
      Overrides for the file search tool.
      - max_num_results: optional number
        
        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 object { score_threshold, ranker }
        
        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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: optional "auto" or "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"
  - FunctionTool object { function, type }
    - function: FunctionDefinition
      - name: string
        
        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 string
        
        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 boolean
        
        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: "function"
      
      The type of tool being defined: function
      - "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.
  - "auto"
    
    auto is the default value
    - "auto"
  - ResponseFormatText object { type }
    
    Default response format. Used to generate text responses.
    - type: "text"
      
      The type of response format being defined. Always text.
      - "text"
  - ResponseFormatJSONObject object { type }
    
    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: "json_object"
      
      The type of response format being defined. Always json_object.
      - "json_object"
  - ResponseFormatJSONSchema object { json_schema, type }
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: object { name, description, schema, strict }
      
      Structured Outputs configuration options, including a JSON Schema.
      - name: string
        
        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 string
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: optional map[unknown]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: optional boolean
        
        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: "json_schema"
      
      The type of response format being defined. Always json_schema.
      - "json_schema"
- temperature: optional number
  
  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 object { code_interpreter, file_search }
  
  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 object { file_ids }
    - file_ids: optional array of string
      
      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 object { vector_store_ids }
    - vector_store_ids: optional array of string
      
      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 number
  
  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

AssistantDeleted object { id, deleted, object }
- id: string
- deleted: boolean
- object: "assistant.deleted"
  - "assistant.deleted"

Assistant Stream Event

AssistantStreamEvent = object { data, event, enabled } or object { data, event } or object { data, event } or 22 more

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.
- object { data, event, enabled }
  
  Occurs when a new thread is created.
  - data: Thread
    
    Represents a thread that contains messages.
    - id: string
      
      The identifier, which can be referenced in API endpoints.
    - created_at: number
      
      The Unix timestamp (in seconds) for when the thread was created.
    - metadata: 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: "thread"
      
      The object type, which is always thread.
      - "thread"
    - tool_resources: object { code_interpreter, file_search }
      
      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 object { file_ids }
        
        file_ids: optional array of string
        
        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 object { vector_store_ids }
        
        vector_store_ids: optional array of string
        
        The vector store attached to this thread. There can be a maximum of 1 vector store attached to the thread.
  - event: "thread.created"
    - "thread.created"
  - enabled: optional boolean
    
    Whether to enable input audio transcription.
- object { data, event }
  
  Occurs when a new run is created.
  - data: Run
    
    Represents an execution run on a thread.
    - id: string
      
      The identifier, which can be referenced in API endpoints.
    - assistant_id: string
      
      The ID of the assistant used for execution of this run.
    - cancelled_at: number
      
      The Unix timestamp (in seconds) for when the run was cancelled.
    - completed_at: number
      
      The Unix timestamp (in seconds) for when the run was completed.
    - created_at: number
      
      The Unix timestamp (in seconds) for when the run was created.
    - expires_at: number
      
      The Unix timestamp (in seconds) for when the run will expire.
    - failed_at: number
      
      The Unix timestamp (in seconds) for when the run failed.
    - incomplete_details: object { reason }
      
      Details on why the run is incomplete. Will be null if the run is not incomplete.
      - reason: optional "max_completion_tokens" or "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: string
      
      The instructions that the assistant used for this run.
    - last_error: object { code, message }
      
      The last error associated with this run. Will be null if there are no errors.
      - code: "server_error" or "rate_limit_exceeded" or "invalid_prompt"
        
        One of server_error, rate_limit_exceeded, or invalid_prompt.
        
        "server_error"
        
        "rate_limit_exceeded"
        
        "invalid_prompt"
      - message: string
        
        A human-readable description of the error.
    - max_completion_tokens: number
      
      The maximum number of completion tokens specified to have been used over the course of the run.
    - max_prompt_tokens: number
      
      The maximum number of prompt tokens specified to have been used over the course of the run.
    - metadata: 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: string
      
      The model that the assistant used for this run.
    - object: "thread.run"
      
      The object type, which is always thread.run.
      - "thread.run"
    - parallel_tool_calls: boolean
      
      Whether to enable parallel function calling during tool use.
    - required_action: object { submit_tool_outputs, type }
      
      Details on the action required to continue the run. Will be null if no action is required.
      - submit_tool_outputs: object { tool_calls }
        
        Details on the tool outputs needed for this run to continue.
        
        tool_calls: array of RequiredActionFunctionToolCall
        
        A list of the relevant tool calls.
        
        id: string
        
        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: object { arguments, name }
        
        The function definition.
        
        arguments: string
        
        The arguments that the model expects you to pass to the function.
        
        name: string
        
        The name of the function.
        
        type: "function"
        
        The type of tool call the output is required for. For now, this is always function.
        
        "function"
      - type: "submit_tool_outputs"
        
        For now, this is always submit_tool_outputs.
        
        "submit_tool_outputs"
    - response_format: 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.
      - "auto"
        
        auto is the default value
        
        "auto"
      - ResponseFormatText object { type }
        
        Default response format. Used to generate text responses.
        
        type: "text"
        
        The type of response format being defined. Always text.
        
        "text"
      - ResponseFormatJSONObject object { type }
        
        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: "json_object"
        
        The type of response format being defined. Always json_object.
        
        "json_object"
      - ResponseFormatJSONSchema object { json_schema, type }
        
        JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
        
        json_schema: object { name, description, schema, strict }
        
        Structured Outputs configuration options, including a JSON Schema.
        
        name: string
        
        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 string
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
        
        schema: optional map[unknown]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
        
        strict: optional boolean
        
        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: "json_schema"
        
        The type of response format being defined. Always json_schema.
        
        "json_schema"
    - started_at: number
      
      The Unix timestamp (in seconds) for when the run was started.
    - status: "queued" or "in_progress" or "requires_action" or 6 more
      
      The status of the run, which can be either queued, in_progress, requires_action, cancelling, cancelled, failed, completed, incomplete, or expired.
      - "queued"
      - "in_progress"
      - "requires_action"
      - "cancelling"
      - "cancelled"
      - "failed"
      - "completed"
      - "incomplete"
      - "expired"
    - thread_id: string
      
      The ID of the thread that was executed on as a part of this run.
    - tool_choice: 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.
      - "none" or "auto" or "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"
      - AssistantToolChoice object { type, function }
        
        Specifies a tool the model should use. Use to force the model to call a specific tool.
        
        type: "function" or "code_interpreter" or "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: string
        
        The name of the function to call.
    - tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool
      
      The list of tools that the assistant used for this run.
      - CodeInterpreterTool object { type }
        
        type: "code_interpreter"
        
        The type of tool being defined: code_interpreter
        
        "code_interpreter"
      - FileSearchTool object { type, file_search }
        
        type: "file_search"
        
        The type of tool being defined: file_search
        
        "file_search"
        
        file_search: optional object { max_num_results, ranking_options }
        
        Overrides for the file search tool.
        
        max_num_results: optional number
        
        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 object { score_threshold, ranker }
        
        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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: optional "auto" or "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"
      - FunctionTool object { function, type }
        
        function: FunctionDefinition
        
        name: string
        
        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 string
        
        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 boolean
        
        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: "function"
        
        The type of tool being defined: function
        
        "function"
    - truncation_strategy: object { type, last_messages }
      
      Controls for how a thread will be truncated prior to the run. Use this to control the initial context window of the run.
      - type: "auto" or "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 number
        
        The number of most recent messages from the thread when constructing the context for the run.
    - usage: object { completion_tokens, prompt_tokens, total_tokens }
      
      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: number
        
        Number of completion tokens used over the course of the run.
      - prompt_tokens: number
        
        Number of prompt tokens used over the course of the run.
      - total_tokens: number
        
        Total number of tokens used (prompt + completion).
    - temperature: optional number
      
      The sampling temperature used for this run. If not set, defaults to 1.
    - top_p: optional number
      
      The nucleus sampling value used for this run. If not set, defaults to 1.
  - event: "thread.run.created"
    - "thread.run.created"
- object { data, event }
  
  Occurs when a run moves to a queued status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.queued"
    - "thread.run.queued"
- object { data, event }
  
  Occurs when a run moves to an in_progress status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.in_progress"
    - "thread.run.in_progress"
- object { data, event }
  
  Occurs when a run moves to a requires_action status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.requires_action"
    - "thread.run.requires_action"
- object { data, event }
  
  Occurs when a run is completed.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.completed"
    - "thread.run.completed"
- object { data, event }
  
  Occurs when a run ends with status incomplete.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.incomplete"
    - "thread.run.incomplete"
- object { data, event }
  
  Occurs when a run fails.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.failed"
    - "thread.run.failed"
- object { data, event }
  
  Occurs when a run moves to a cancelling status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.cancelling"
    - "thread.run.cancelling"
- object { data, event }
  
  Occurs when a run is cancelled.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.cancelled"
    - "thread.run.cancelled"
- object { data, event }
  
  Occurs when a run expires.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.expired"
    - "thread.run.expired"
- object { data, event }
  
  Occurs when a run step is created.
  - data: RunStep
    
    Represents a step in execution of a run.
    - id: string
      
      The identifier of the run step, which can be referenced in API endpoints.
    - assistant_id: string
      
      The ID of the assistant associated with the run step.
    - cancelled_at: number
      
      The Unix timestamp (in seconds) for when the run step was cancelled.
    - completed_at: number
      
      The Unix timestamp (in seconds) for when the run step completed.
    - created_at: number
      
      The Unix timestamp (in seconds) for when the run step was created.
    - expired_at: number
      
      The Unix timestamp (in seconds) for when the run step expired. A step is considered expired if the parent run is expired.
    - failed_at: number
      
      The Unix timestamp (in seconds) for when the run step failed.
    - last_error: object { code, message }
      
      The last error associated with this run step. Will be null if there are no errors.
      - code: "server_error" or "rate_limit_exceeded"
        
        One of server_error or rate_limit_exceeded.
        
        "server_error"
        
        "rate_limit_exceeded"
      - message: string
        
        A human-readable description of the error.
    - metadata: 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: "thread.run.step"
      
      The object type, which is always thread.run.step.
      - "thread.run.step"
    - run_id: string
      
      The ID of the run that this run step is a part of.
    - status: "in_progress" or "cancelled" or "failed" or 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: MessageCreationStepDetails or ToolCallsStepDetails
      
      The details of the run step.
      - MessageCreationStepDetails object { message_creation, type }
        
        Details of the message creation by the run step.
        
        message_creation: object { message_id }
        
        message_id: string
        
        The ID of the message that was created by this run step.
        
        type: "message_creation"
        
        Always message_creation.
        
        "message_creation"
      - ToolCallsStepDetails object { tool_calls, type }
        
        Details of the tool call.
        
        tool_calls: array of CodeInterpreterToolCall or FileSearchToolCall or FunctionToolCall
        
        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.
        
        CodeInterpreterToolCall object { id, code_interpreter, type }
        
        Details of the Code Interpreter tool call the run step was involved in.
        
        id: string
        
        The ID of the tool call.
        
        code_interpreter: object { input, outputs }
        
        The Code Interpreter tool call definition.
        
        input: string
        
        The input to the Code Interpreter tool call.
        
        outputs: array of object { logs, type } or object { image, type }
        
        The outputs from the Code Interpreter tool call. Code Interpreter can output one or more items, including text (logs) or images (image). Each of these are represented by a different object type.
        
        CodeInterpreterLogOutput object { logs, type }
        
        Text output from the Code Interpreter tool call as part of a run step.
        
        logs: string
        
        The text output from the Code Interpreter tool call.
        
        type: "logs"
        
        Always logs.
        
        "logs"
        
        CodeInterpreterImageOutput object { image, type }
        
        image: object { file_id }
        
        file_id: string
        
        The file ID of the image.
        
        type: "image"
        
        Always image.
        
        "image"
        
        type: "code_interpreter"
        
        The type of tool call. This is always going to be code_interpreter for this type of tool call.
        
        "code_interpreter"
        
        FileSearchToolCall object { id, file_search, type }
        
        id: string
        
        The ID of the tool call object.
        
        file_search: object { ranking_options, results }
        
        For now, this is always going to be an empty object.
        
        ranking_options: optional object { ranker, score_threshold }
        
        The ranking options for the file search.
        
        ranker: "auto" or "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"
        
        score_threshold: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        results: optional array of object { file_id, file_name, score, content }
        
        The results of the file search.
        
        file_id: string
        
        The ID of the file that result was found in.
        
        file_name: string
        
        The name of the file that result was found in.
        
        score: number
        
        The score of the result. All values must be a floating point number between 0 and 1.
        
        content: optional array of object { text, type }
        
        The content of the result that was found. The content is only included if requested via the include query parameter.
        
        text: optional string
        
        The text content of the file.
        
        type: optional "text"
        
        The type of the content.
        
        "text"
        
        type: "file_search"
        
        The type of tool call. This is always going to be file_search for this type of tool call.
        
        "file_search"
        
        FunctionToolCall object { id, function, type }
        
        id: string
        
        The ID of the tool call object.
        
        function: object { arguments, name, output }
        
        The definition of the function that was called.
        
        arguments: string
        
        The arguments passed to the function.
        
        name: string
        
        The name of the function.
        
        output: string
        
        The output of the function. This will be null if the outputs have not been submitted yet.
        
        type: "function"
        
        The type of tool call. This is always going to be function for this type of tool call.
        
        "function"
        
        type: "tool_calls"
        
        Always tool_calls.
        
        "tool_calls"
    - thread_id: string
      
      The ID of the thread that was run.
    - type: "message_creation" or "tool_calls"
      
      The type of run step, which can be either message_creation or tool_calls.
      - "message_creation"
      - "tool_calls"
    - usage: object { completion_tokens, prompt_tokens, total_tokens }
      
      Usage statistics related to the run step. This value will be null while the run step's status is in_progress.
      - completion_tokens: number
        
        Number of completion tokens used over the course of the run step.
      - prompt_tokens: number
        
        Number of prompt tokens used over the course of the run step.
      - total_tokens: number
        
        Total number of tokens used (prompt + completion).
  - event: "thread.run.step.created"
    - "thread.run.step.created"
- object { data, event }
  
  Occurs when a run step moves to an in_progress state.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.in_progress"
    - "thread.run.step.in_progress"
- object { data, event }
  
  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: string
      
      The identifier of the run step, which can be referenced in API endpoints.
    - delta: object { step_details }
      
      The delta containing the fields that have changed on the run step.
      - step_details: optional RunStepDeltaMessageDelta or ToolCallDeltaObject
        
        The details of the run step.
        
        RunStepDeltaMessageDelta object { type, message_creation }
        
        Details of the message creation by the run step.
        
        type: "message_creation"
        
        Always message_creation.
        
        "message_creation"
        
        message_creation: optional object { message_id }
        
        message_id: optional string
        
        The ID of the message that was created by this run step.
        
        ToolCallDeltaObject object { type, tool_calls }
        
        Details of the tool call.
        
        type: "tool_calls"
        
        Always tool_calls.
        
        "tool_calls"
        
        tool_calls: optional array of CodeInterpreterToolCallDelta or FileSearchToolCallDelta or FunctionToolCallDelta
        
        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.
        
        CodeInterpreterToolCallDelta object { index, type, id, code_interpreter }
        
        Details of the Code Interpreter tool call the run step was involved in.
        
        index: number
        
        The index of the tool call in the tool calls array.
        
        type: "code_interpreter"
        
        The type of tool call. This is always going to be code_interpreter for this type of tool call.
        
        "code_interpreter"
        
        id: optional string
        
        The ID of the tool call.
        
        code_interpreter: optional object { input, outputs }
        
        The Code Interpreter tool call definition.
        
        input: optional string
        
        The input to the Code Interpreter tool call.
        
        outputs: optional array of CodeInterpreterLogs or CodeInterpreterOutputImage
        
        The outputs from the Code Interpreter tool call. Code Interpreter can output one or more items, including text (logs) or images (image). Each of these are represented by a different object type.
        
        CodeInterpreterLogs object { index, type, logs }
        
        Text output from the Code Interpreter tool call as part of a run step.
        
        index: number
        
        The index of the output in the outputs array.
        
        type: "logs"
        
        Always logs.
        
        "logs"
        
        logs: optional string
        
        The text output from the Code Interpreter tool call.
        
        CodeInterpreterOutputImage object { index, type, image }
        
        index: number
        
        The index of the output in the outputs array.
        
        type: "image"
        
        Always image.
        
        "image"
        
        image: optional object { file_id }
        
        file_id: optional string
        
        The file ID of the image.
        
        FileSearchToolCallDelta object { file_search, index, type, id }
        
        file_search: unknown
        
        For now, this is always going to be an empty object.
        
        index: number
        
        The index of the tool call in the tool calls array.
        
        type: "file_search"
        
        The type of tool call. This is always going to be file_search for this type of tool call.
        
        "file_search"
        
        id: optional string
        
        The ID of the tool call object.
        
        FunctionToolCallDelta object { index, type, id, function }
        
        index: number
        
        The index of the tool call in the tool calls array.
        
        type: "function"
        
        The type of tool call. This is always going to be function for this type of tool call.
        
        "function"
        
        id: optional string
        
        The ID of the tool call object.
        
        function: optional object { arguments, name, output }
        
        The definition of the function that was called.
        
        arguments: optional string
        
        The arguments passed to the function.
        
        name: optional string
        
        The name of the function.
        
        output: optional string
        
        The output of the function. This will be null if the outputs have not been submitted yet.
    - object: "thread.run.step.delta"
      
      The object type, which is always thread.run.step.delta.
      - "thread.run.step.delta"
  - event: "thread.run.step.delta"
    - "thread.run.step.delta"
- object { data, event }
  
  Occurs when a run step is completed.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.completed"
    - "thread.run.step.completed"
- object { data, event }
  
  Occurs when a run step fails.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.failed"
    - "thread.run.step.failed"
- object { data, event }
  
  Occurs when a run step is cancelled.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.cancelled"
    - "thread.run.step.cancelled"
- object { data, event }
  
  Occurs when a run step expires.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.expired"
    - "thread.run.step.expired"
- object { data, event }
  
  Occurs when a message is created.
  - data: Message
    
    Represents a message within a thread.
    - id: string
      
      The identifier, which can be referenced in API endpoints.
    - assistant_id: string
      
      If applicable, the ID of the assistant that authored this message.
    - attachments: array of object { file_id, tools }
      
      A list of files attached to the message, and the tools they were added to.
      - file_id: optional string
        
        The ID of the file to attach to the message.
      - tools: optional array of CodeInterpreterTool or object { type }
        
        The tools to add this file to.
        
        CodeInterpreterTool object { type }
        
        FileSearchTool object { type }
        
        type: "file_search"
        
        The type of tool being defined: file_search
        
        "file_search"
    - completed_at: number
      
      The Unix timestamp (in seconds) for when the message was completed.
    - content: array of ImageFileContentBlock or ImageURLContentBlock or TextContentBlock or RefusalContentBlock
      
      The content of the message in array of text and/or images.
      - ImageFileContentBlock object { image_file, type }
        
        References an image File in the content of a message.
        
        image_file: ImageFile
        
        file_id: string
        
        The File ID of the image in the message content. Set purpose="vision" when uploading the File if you need to later display the file content.
        
        detail: optional "auto" or "low" or "high"
        
        Specifies the detail level of the image if specified by the user. low uses fewer tokens, you can opt in to high resolution using high.
        
        "auto"
        
        "low"
        
        "high"
        
        type: "image_file"
        
        Always image_file.
        
        "image_file"
      - ImageURLContentBlock object { image_url, type }
        
        References an image URL in the content of a message.
        
        image_url: ImageURL
        
        url: string
        
        The external URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.
        
        detail: optional "auto" or "low" or "high"
        
        Specifies the detail level of the image. low uses fewer tokens, you can opt in to high resolution using high. Default value is auto
        
        "auto"
        
        "low"
        
        "high"
        
        type: "image_url"
        
        The type of the content part.
        
        "image_url"
      - TextContentBlock object { text, type }
        
        The text content that is part of a message.
        
        text: Text
        
        annotations: array of FileCitationAnnotation or FilePathAnnotation
        
        FileCitationAnnotation object { end_index, file_citation, start_index, 2 more }
        
        A citation within the message that points to a specific quote from a specific File associated with the assistant or the message. Generated when the assistant uses the "file_search" tool to search files.
        
        end_index: number
        
        file_citation: object { file_id }
        
        file_id: string
        
        The ID of the specific File the citation is from.
        
        start_index: number
        
        text: string
        
        The text in the message content that needs to be replaced.
        
        type: "file_citation"
        
        Always file_citation.
        
        "file_citation"
        
        FilePathAnnotation object { end_index, file_path, start_index, 2 more }
        
        A URL for the file that's generated when the assistant used the code_interpreter tool to generate a file.
        
        end_index: number
        
        file_path: object { file_id }
        
        file_id: string
        
        The ID of the file that was generated.
        
        start_index: number
        
        text: string
        
        The text in the message content that needs to be replaced.
        
        type: "file_path"
        
        Always file_path.
        
        "file_path"
        
        value: string
        
        The data that makes up the text.
        
        type: "text"
        
        Always text.
        
        "text"
      - RefusalContentBlock object { refusal, type }
        
        The refusal content generated by the assistant.
        
        refusal: string
        
        type: "refusal"
        
        Always refusal.
        
        "refusal"
    - created_at: number
      
      The Unix timestamp (in seconds) for when the message was created.
    - incomplete_at: number
      
      The Unix timestamp (in seconds) for when the message was marked as incomplete.
    - incomplete_details: object { reason }
      
      On an incomplete message, details about why the message is incomplete.
      - reason: "content_filter" or "max_tokens" or "run_cancelled" or 2 more
        
        The reason the message is incomplete.
        
        "content_filter"
        
        "max_tokens"
        
        "run_cancelled"
        
        "run_expired"
        
        "run_failed"
    - metadata: 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: "thread.message"
      
      The object type, which is always thread.message.
      - "thread.message"
    - role: "user" or "assistant"
      
      The entity that produced the message. One of user or assistant.
      - "user"
      - "assistant"
    - run_id: string
      
      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: "in_progress" or "incomplete" or "completed"
      
      The status of the message, which can be either in_progress, incomplete, or completed.
      - "in_progress"
      - "incomplete"
      - "completed"
    - thread_id: string
      
      The thread ID that this message belongs to.
  - event: "thread.message.created"
    - "thread.message.created"
- object { data, event }
  
  Occurs when a message moves to an in_progress state.
  - data: Message
    
    Represents a message within a thread.
  - event: "thread.message.in_progress"
    - "thread.message.in_progress"
- object { data, event }
  
  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: string
      
      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 array of ImageFileDeltaBlock or TextDeltaBlock or RefusalDeltaBlock or ImageURLDeltaBlock
        
        The content of the message in array of text and/or images.
        
        ImageFileDeltaBlock object { index, type, image_file }
        
        References an image File in the content of a message.
        
        index: number
        
        The index of the content part in the message.
        
        type: "image_file"
        
        Always image_file.
        
        "image_file"
        
        image_file: optional ImageFileDelta
        
        detail: optional "auto" or "low" or "high"
        
        Specifies the detail level of the image if specified by the user. low uses fewer tokens, you can opt in to high resolution using high.
        
        "auto"
        
        "low"
        
        "high"
        
        file_id: optional string
        
        The File ID of the image in the message content. Set purpose="vision" when uploading the File if you need to later display the file content.
        
        TextDeltaBlock object { index, type, text }
        
        The text content that is part of a message.
        
        index: number
        
        The index of the content part in the message.
        
        type: "text"
        
        Always text.
        
        "text"
        
        text: optional TextDelta
        
        annotations: optional array of FileCitationDeltaAnnotation or FilePathDeltaAnnotation
        
        FileCitationDeltaAnnotation object { index, type, end_index, 3 more }
        
        A citation within the message that points to a specific quote from a specific File associated with the assistant or the message. Generated when the assistant uses the "file_search" tool to search files.
        
        index: number
        
        The index of the annotation in the text content part.
        
        type: "file_citation"
        
        Always file_citation.
        
        "file_citation"
        
        end_index: optional number
        
        file_citation: optional object { file_id, quote }
        
        file_id: optional string
        
        The ID of the specific File the citation is from.
        
        quote: optional string
        
        The specific quote in the file.
        
        start_index: optional number
        
        text: optional string
        
        The text in the message content that needs to be replaced.
        
        FilePathDeltaAnnotation object { index, type, end_index, 3 more }
        
        A URL for the file that's generated when the assistant used the code_interpreter tool to generate a file.
        
        index: number
        
        The index of the annotation in the text content part.
        
        type: "file_path"
        
        Always file_path.
        
        "file_path"
        
        end_index: optional number
        
        file_path: optional object { file_id }
        
        file_id: optional string
        
        The ID of the file that was generated.
        
        start_index: optional number
        
        text: optional string
        
        The text in the message content that needs to be replaced.
        
        value: optional string
        
        The data that makes up the text.
        
        RefusalDeltaBlock object { index, type, refusal }
        
        The refusal content that is part of a message.
        
        index: number
        
        The index of the refusal part in the message.
        
        type: "refusal"
        
        Always refusal.
        
        "refusal"
        
        refusal: optional string
        
        ImageURLDeltaBlock object { index, type, image_url }
        
        References an image URL in the content of a message.
        
        index: number
        
        The index of the content part in the message.
        
        type: "image_url"
        
        Always image_url.
        
        "image_url"
        
        image_url: optional ImageURLDelta
        
        detail: optional "auto" or "low" or "high"
        
        Specifies the detail level of the image. low uses fewer tokens, you can opt in to high resolution using high.
        
        "auto"
        
        "low"
        
        "high"
        
        url: optional string
        
        The URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.
      - role: optional "user" or "assistant"
        
        The entity that produced the message. One of user or assistant.
        
        "user"
        
        "assistant"
    - object: "thread.message.delta"
      
      The object type, which is always thread.message.delta.
      - "thread.message.delta"
  - event: "thread.message.delta"
    - "thread.message.delta"
- object { data, event }
  
  Occurs when a message is completed.
  - data: Message
    
    Represents a message within a thread.
  - event: "thread.message.completed"
    - "thread.message.completed"
- object { data, event }
  
  Occurs when a message ends before it is completed.
  - data: Message
    
    Represents a message within a thread.
  - event: "thread.message.incomplete"
    - "thread.message.incomplete"
- ErrorEvent object { data, event }
  
  Occurs when an error occurs. This can happen due to an internal server error or a timeout.
  - data: ErrorObject
    - code: string
    - message: string
    - param: string
    - type: string
  - event: "error"
    - "error"
- DoneEvent object { data, event }
  
  Occurs when a stream ends.
  - data: "[DONE]"
    - "[DONE]"
  - event: "done"
    - "done"

Code Interpreter Tool

CodeInterpreterTool object { type }
- type: "code_interpreter"
  
  The type of tool being defined: code_interpreter
  - "code_interpreter"

File Search Tool

FileSearchTool object { type, file_search }
- type: "file_search"
  
  The type of tool being defined: file_search
  - "file_search"
- file_search: optional object { max_num_results, ranking_options }
  
  Overrides for the file search tool.
  - max_num_results: optional number
    
    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 object { score_threshold, ranker }
    
    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: number
      
      The score threshold for the file search. All values must be a floating point number between 0 and 1.
    - ranker: optional "auto" or "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

FunctionTool object { function, type }
- function: FunctionDefinition
  - name: string
    
    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 string
    
    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 boolean
    
    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: "function"
  
  The type of tool being defined: function
  - "function"

Message Stream Event

MessageStreamEvent = object { data, event } or object { data, event } or object { data, event } or 2 more

Occurs when a message is created.
- object { data, event }
  
  Occurs when a message is created.
  - data: Message
    
    Represents a message within a thread.
    - id: string
      
      The identifier, which can be referenced in API endpoints.
    - assistant_id: string
      
      If applicable, the ID of the assistant that authored this message.
    - attachments: array of object { file_id, tools }
      
      A list of files attached to the message, and the tools they were added to.
      - file_id: optional string
        
        The ID of the file to attach to the message.
      - tools: optional array of CodeInterpreterTool or object { type }
        
        The tools to add this file to.
        
        CodeInterpreterTool object { type }
        
        type: "code_interpreter"
        
        The type of tool being defined: code_interpreter
        
        "code_interpreter"
        
        FileSearchTool object { type }
        
        type: "file_search"
        
        The type of tool being defined: file_search
        
        "file_search"
    - completed_at: number
      
      The Unix timestamp (in seconds) for when the message was completed.
    - content: array of ImageFileContentBlock or ImageURLContentBlock or TextContentBlock or RefusalContentBlock
      
      The content of the message in array of text and/or images.
      - ImageFileContentBlock object { image_file, type }
        
        References an image File in the content of a message.
        
        image_file: ImageFile
        
        file_id: string
        
        The File ID of the image in the message content. Set purpose="vision" when uploading the File if you need to later display the file content.
        
        detail: optional "auto" or "low" or "high"
        
        Specifies the detail level of the image if specified by the user. low uses fewer tokens, you can opt in to high resolution using high.
        
        "auto"
        
        "low"
        
        "high"
        
        type: "image_file"
        
        Always image_file.
        
        "image_file"
      - ImageURLContentBlock object { image_url, type }
        
        References an image URL in the content of a message.
        
        image_url: ImageURL
        
        url: string
        
        The external URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.
        
        detail: optional "auto" or "low" or "high"
        
        Specifies the detail level of the image. low uses fewer tokens, you can opt in to high resolution using high. Default value is auto
        
        "auto"
        
        "low"
        
        "high"
        
        type: "image_url"
        
        The type of the content part.
        
        "image_url"
      - TextContentBlock object { text, type }
        
        The text content that is part of a message.
        
        text: Text
        
        annotations: array of FileCitationAnnotation or FilePathAnnotation
        
        FileCitationAnnotation object { end_index, file_citation, start_index, 2 more }
        
        A citation within the message that points to a specific quote from a specific File associated with the assistant or the message. Generated when the assistant uses the "file_search" tool to search files.
        
        end_index: number
        
        file_citation: object { file_id }
        
        file_id: string
        
        The ID of the specific File the citation is from.
        
        start_index: number
        
        text: string
        
        The text in the message content that needs to be replaced.
        
        type: "file_citation"
        
        Always file_citation.
        
        "file_citation"
        
        FilePathAnnotation object { end_index, file_path, start_index, 2 more }
        
        A URL for the file that's generated when the assistant used the code_interpreter tool to generate a file.
        
        end_index: number
        
        file_path: object { file_id }
        
        file_id: string
        
        The ID of the file that was generated.
        
        start_index: number
        
        text: string
        
        The text in the message content that needs to be replaced.
        
        type: "file_path"
        
        Always file_path.
        
        "file_path"
        
        value: string
        
        The data that makes up the text.
        
        type: "text"
        
        Always text.
        
        "text"
      - RefusalContentBlock object { refusal, type }
        
        The refusal content generated by the assistant.
        
        refusal: string
        
        type: "refusal"
        
        Always refusal.
        
        "refusal"
    - created_at: number
      
      The Unix timestamp (in seconds) for when the message was created.
    - incomplete_at: number
      
      The Unix timestamp (in seconds) for when the message was marked as incomplete.
    - incomplete_details: object { reason }
      
      On an incomplete message, details about why the message is incomplete.
      - reason: "content_filter" or "max_tokens" or "run_cancelled" or 2 more
        
        The reason the message is incomplete.
        
        "content_filter"
        
        "max_tokens"
        
        "run_cancelled"
        
        "run_expired"
        
        "run_failed"
    - metadata: 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: "thread.message"
      
      The object type, which is always thread.message.
      - "thread.message"
    - role: "user" or "assistant"
      
      The entity that produced the message. One of user or assistant.
      - "user"
      - "assistant"
    - run_id: string
      
      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: "in_progress" or "incomplete" or "completed"
      
      The status of the message, which can be either in_progress, incomplete, or completed.
      - "in_progress"
      - "incomplete"
      - "completed"
    - thread_id: string
      
      The thread ID that this message belongs to.
  - event: "thread.message.created"
    - "thread.message.created"
- object { data, event }
  
  Occurs when a message moves to an in_progress state.
  - data: Message
    
    Represents a message within a thread.
  - event: "thread.message.in_progress"
    - "thread.message.in_progress"
- object { data, event }
  
  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: string
      
      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 array of ImageFileDeltaBlock or TextDeltaBlock or RefusalDeltaBlock or ImageURLDeltaBlock
        
        The content of the message in array of text and/or images.
        
        ImageFileDeltaBlock object { index, type, image_file }
        
        References an image File in the content of a message.
        
        index: number
        
        The index of the content part in the message.
        
        type: "image_file"
        
        Always image_file.
        
        "image_file"
        
        image_file: optional ImageFileDelta
        
        detail: optional "auto" or "low" or "high"
        
        Specifies the detail level of the image if specified by the user. low uses fewer tokens, you can opt in to high resolution using high.
        
        "auto"
        
        "low"
        
        "high"
        
        file_id: optional string
        
        The File ID of the image in the message content. Set purpose="vision" when uploading the File if you need to later display the file content.
        
        TextDeltaBlock object { index, type, text }
        
        The text content that is part of a message.
        
        index: number
        
        The index of the content part in the message.
        
        type: "text"
        
        Always text.
        
        "text"
        
        text: optional TextDelta
        
        annotations: optional array of FileCitationDeltaAnnotation or FilePathDeltaAnnotation
        
        FileCitationDeltaAnnotation object { index, type, end_index, 3 more }
        
        A citation within the message that points to a specific quote from a specific File associated with the assistant or the message. Generated when the assistant uses the "file_search" tool to search files.
        
        index: number
        
        The index of the annotation in the text content part.
        
        type: "file_citation"
        
        Always file_citation.
        
        "file_citation"
        
        end_index: optional number
        
        file_citation: optional object { file_id, quote }
        
        file_id: optional string
        
        The ID of the specific File the citation is from.
        
        quote: optional string
        
        The specific quote in the file.
        
        start_index: optional number
        
        text: optional string
        
        The text in the message content that needs to be replaced.
        
        FilePathDeltaAnnotation object { index, type, end_index, 3 more }
        
        A URL for the file that's generated when the assistant used the code_interpreter tool to generate a file.
        
        index: number
        
        The index of the annotation in the text content part.
        
        type: "file_path"
        
        Always file_path.
        
        "file_path"
        
        end_index: optional number
        
        file_path: optional object { file_id }
        
        file_id: optional string
        
        The ID of the file that was generated.
        
        start_index: optional number
        
        text: optional string
        
        The text in the message content that needs to be replaced.
        
        value: optional string
        
        The data that makes up the text.
        
        RefusalDeltaBlock object { index, type, refusal }
        
        The refusal content that is part of a message.
        
        index: number
        
        The index of the refusal part in the message.
        
        type: "refusal"
        
        Always refusal.
        
        "refusal"
        
        refusal: optional string
        
        ImageURLDeltaBlock object { index, type, image_url }
        
        References an image URL in the content of a message.
        
        index: number
        
        The index of the content part in the message.
        
        type: "image_url"
        
        Always image_url.
        
        "image_url"
        
        image_url: optional ImageURLDelta
        
        detail: optional "auto" or "low" or "high"
        
        Specifies the detail level of the image. low uses fewer tokens, you can opt in to high resolution using high.
        
        "auto"
        
        "low"
        
        "high"
        
        url: optional string
        
        The URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.
      - role: optional "user" or "assistant"
        
        The entity that produced the message. One of user or assistant.
        
        "user"
        
        "assistant"
    - object: "thread.message.delta"
      
      The object type, which is always thread.message.delta.
      - "thread.message.delta"
  - event: "thread.message.delta"
    - "thread.message.delta"
- object { data, event }
  
  Occurs when a message is completed.
  - data: Message
    
    Represents a message within a thread.
  - event: "thread.message.completed"
    - "thread.message.completed"
- object { data, event }
  
  Occurs when a message ends before it is completed.
  - data: Message
    
    Represents a message within a thread.
  - event: "thread.message.incomplete"
    - "thread.message.incomplete"

Run Step Stream Event

RunStepStreamEvent = object { data, event } or object { data, event } or object { data, event } or 4 more

Occurs when a run step is created.
- object { data, event }
  
  Occurs when a run step is created.
  - data: RunStep
    
    Represents a step in execution of a run.
    - id: string
      
      The identifier of the run step, which can be referenced in API endpoints.
    - assistant_id: string
      
      The ID of the assistant associated with the run step.
    - cancelled_at: number
      
      The Unix timestamp (in seconds) for when the run step was cancelled.
    - completed_at: number
      
      The Unix timestamp (in seconds) for when the run step completed.
    - created_at: number
      
      The Unix timestamp (in seconds) for when the run step was created.
    - expired_at: number
      
      The Unix timestamp (in seconds) for when the run step expired. A step is considered expired if the parent run is expired.
    - failed_at: number
      
      The Unix timestamp (in seconds) for when the run step failed.
    - last_error: object { code, message }
      
      The last error associated with this run step. Will be null if there are no errors.
      - code: "server_error" or "rate_limit_exceeded"
        
        One of server_error or rate_limit_exceeded.
        
        "server_error"
        
        "rate_limit_exceeded"
      - message: string
        
        A human-readable description of the error.
    - metadata: 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: "thread.run.step"
      
      The object type, which is always thread.run.step.
      - "thread.run.step"
    - run_id: string
      
      The ID of the run that this run step is a part of.
    - status: "in_progress" or "cancelled" or "failed" or 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: MessageCreationStepDetails or ToolCallsStepDetails
      
      The details of the run step.
      - MessageCreationStepDetails object { message_creation, type }
        
        Details of the message creation by the run step.
        
        message_creation: object { message_id }
        
        message_id: string
        
        The ID of the message that was created by this run step.
        
        type: "message_creation"
        
        Always message_creation.
        
        "message_creation"
      - ToolCallsStepDetails object { tool_calls, type }
        
        Details of the tool call.
        
        tool_calls: array of CodeInterpreterToolCall or FileSearchToolCall or FunctionToolCall
        
        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.
        
        CodeInterpreterToolCall object { id, code_interpreter, type }
        
        Details of the Code Interpreter tool call the run step was involved in.
        
        id: string
        
        The ID of the tool call.
        
        code_interpreter: object { input, outputs }
        
        The Code Interpreter tool call definition.
        
        input: string
        
        The input to the Code Interpreter tool call.
        
        outputs: array of object { logs, type } or object { image, type }
        
        The outputs from the Code Interpreter tool call. Code Interpreter can output one or more items, including text (logs) or images (image). Each of these are represented by a different object type.
        
        CodeInterpreterLogOutput object { logs, type }
        
        Text output from the Code Interpreter tool call as part of a run step.
        
        logs: string
        
        The text output from the Code Interpreter tool call.
        
        type: "logs"
        
        Always logs.
        
        "logs"
        
        CodeInterpreterImageOutput object { image, type }
        
        image: object { file_id }
        
        file_id: string
        
        The file ID of the image.
        
        type: "image"
        
        Always image.
        
        "image"
        
        type: "code_interpreter"
        
        The type of tool call. This is always going to be code_interpreter for this type of tool call.
        
        "code_interpreter"
        
        FileSearchToolCall object { id, file_search, type }
        
        id: string
        
        The ID of the tool call object.
        
        file_search: object { ranking_options, results }
        
        For now, this is always going to be an empty object.
        
        ranking_options: optional object { ranker, score_threshold }
        
        The ranking options for the file search.
        
        ranker: "auto" or "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"
        
        score_threshold: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        results: optional array of object { file_id, file_name, score, content }
        
        The results of the file search.
        
        file_id: string
        
        The ID of the file that result was found in.
        
        file_name: string
        
        The name of the file that result was found in.
        
        score: number
        
        The score of the result. All values must be a floating point number between 0 and 1.
        
        content: optional array of object { text, type }
        
        The content of the result that was found. The content is only included if requested via the include query parameter.
        
        text: optional string
        
        The text content of the file.
        
        type: optional "text"
        
        The type of the content.
        
        "text"
        
        type: "file_search"
        
        The type of tool call. This is always going to be file_search for this type of tool call.
        
        "file_search"
        
        FunctionToolCall object { id, function, type }
        
        id: string
        
        The ID of the tool call object.
        
        function: object { arguments, name, output }
        
        The definition of the function that was called.
        
        arguments: string
        
        The arguments passed to the function.
        
        name: string
        
        The name of the function.
        
        output: string
        
        The output of the function. This will be null if the outputs have not been submitted yet.
        
        type: "function"
        
        The type of tool call. This is always going to be function for this type of tool call.
        
        "function"
        
        type: "tool_calls"
        
        Always tool_calls.
        
        "tool_calls"
    - thread_id: string
      
      The ID of the thread that was run.
    - type: "message_creation" or "tool_calls"
      
      The type of run step, which can be either message_creation or tool_calls.
      - "message_creation"
      - "tool_calls"
    - usage: object { completion_tokens, prompt_tokens, total_tokens }
      
      Usage statistics related to the run step. This value will be null while the run step's status is in_progress.
      - completion_tokens: number
        
        Number of completion tokens used over the course of the run step.
      - prompt_tokens: number
        
        Number of prompt tokens used over the course of the run step.
      - total_tokens: number
        
        Total number of tokens used (prompt + completion).
  - event: "thread.run.step.created"
    - "thread.run.step.created"
- object { data, event }
  
  Occurs when a run step moves to an in_progress state.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.in_progress"
    - "thread.run.step.in_progress"
- object { data, event }
  
  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: string
      
      The identifier of the run step, which can be referenced in API endpoints.
    - delta: object { step_details }
      
      The delta containing the fields that have changed on the run step.
      - step_details: optional RunStepDeltaMessageDelta or ToolCallDeltaObject
        
        The details of the run step.
        
        RunStepDeltaMessageDelta object { type, message_creation }
        
        Details of the message creation by the run step.
        
        type: "message_creation"
        
        Always message_creation.
        
        "message_creation"
        
        message_creation: optional object { message_id }
        
        message_id: optional string
        
        The ID of the message that was created by this run step.
        
        ToolCallDeltaObject object { type, tool_calls }
        
        Details of the tool call.
        
        type: "tool_calls"
        
        Always tool_calls.
        
        "tool_calls"
        
        tool_calls: optional array of CodeInterpreterToolCallDelta or FileSearchToolCallDelta or FunctionToolCallDelta
        
        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.
        
        CodeInterpreterToolCallDelta object { index, type, id, code_interpreter }
        
        Details of the Code Interpreter tool call the run step was involved in.
        
        index: number
        
        The index of the tool call in the tool calls array.
        
        type: "code_interpreter"
        
        The type of tool call. This is always going to be code_interpreter for this type of tool call.
        
        "code_interpreter"
        
        id: optional string
        
        The ID of the tool call.
        
        code_interpreter: optional object { input, outputs }
        
        The Code Interpreter tool call definition.
        
        input: optional string
        
        The input to the Code Interpreter tool call.
        
        outputs: optional array of CodeInterpreterLogs or CodeInterpreterOutputImage
        
        The outputs from the Code Interpreter tool call. Code Interpreter can output one or more items, including text (logs) or images (image). Each of these are represented by a different object type.
        
        CodeInterpreterLogs object { index, type, logs }
        
        Text output from the Code Interpreter tool call as part of a run step.
        
        index: number
        
        The index of the output in the outputs array.
        
        type: "logs"
        
        Always logs.
        
        "logs"
        
        logs: optional string
        
        The text output from the Code Interpreter tool call.
        
        CodeInterpreterOutputImage object { index, type, image }
        
        index: number
        
        The index of the output in the outputs array.
        
        type: "image"
        
        Always image.
        
        "image"
        
        image: optional object { file_id }
        
        file_id: optional string
        
        The file ID of the image.
        
        FileSearchToolCallDelta object { file_search, index, type, id }
        
        file_search: unknown
        
        For now, this is always going to be an empty object.
        
        index: number
        
        The index of the tool call in the tool calls array.
        
        type: "file_search"
        
        The type of tool call. This is always going to be file_search for this type of tool call.
        
        "file_search"
        
        id: optional string
        
        The ID of the tool call object.
        
        FunctionToolCallDelta object { index, type, id, function }
        
        index: number
        
        The index of the tool call in the tool calls array.
        
        type: "function"
        
        The type of tool call. This is always going to be function for this type of tool call.
        
        "function"
        
        id: optional string
        
        The ID of the tool call object.
        
        function: optional object { arguments, name, output }
        
        The definition of the function that was called.
        
        arguments: optional string
        
        The arguments passed to the function.
        
        name: optional string
        
        The name of the function.
        
        output: optional string
        
        The output of the function. This will be null if the outputs have not been submitted yet.
    - object: "thread.run.step.delta"
      
      The object type, which is always thread.run.step.delta.
      - "thread.run.step.delta"
  - event: "thread.run.step.delta"
    - "thread.run.step.delta"
- object { data, event }
  
  Occurs when a run step is completed.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.completed"
    - "thread.run.step.completed"
- object { data, event }
  
  Occurs when a run step fails.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.failed"
    - "thread.run.step.failed"
- object { data, event }
  
  Occurs when a run step is cancelled.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.cancelled"
    - "thread.run.step.cancelled"
- object { data, event }
  
  Occurs when a run step expires.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: "thread.run.step.expired"
    - "thread.run.step.expired"

Run Stream Event

RunStreamEvent = object { data, event } or object { data, event } or object { data, event } or 7 more

Occurs when a new run is created.
- object { data, event }
  
  Occurs when a new run is created.
  - data: Run
    
    Represents an execution run on a thread.
    - id: string
      
      The identifier, which can be referenced in API endpoints.
    - assistant_id: string
      
      The ID of the assistant used for execution of this run.
    - cancelled_at: number
      
      The Unix timestamp (in seconds) for when the run was cancelled.
    - completed_at: number
      
      The Unix timestamp (in seconds) for when the run was completed.
    - created_at: number
      
      The Unix timestamp (in seconds) for when the run was created.
    - expires_at: number
      
      The Unix timestamp (in seconds) for when the run will expire.
    - failed_at: number
      
      The Unix timestamp (in seconds) for when the run failed.
    - incomplete_details: object { reason }
      
      Details on why the run is incomplete. Will be null if the run is not incomplete.
      - reason: optional "max_completion_tokens" or "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: string
      
      The instructions that the assistant used for this run.
    - last_error: object { code, message }
      
      The last error associated with this run. Will be null if there are no errors.
      - code: "server_error" or "rate_limit_exceeded" or "invalid_prompt"
        
        One of server_error, rate_limit_exceeded, or invalid_prompt.
        
        "server_error"
        
        "rate_limit_exceeded"
        
        "invalid_prompt"
      - message: string
        
        A human-readable description of the error.
    - max_completion_tokens: number
      
      The maximum number of completion tokens specified to have been used over the course of the run.
    - max_prompt_tokens: number
      
      The maximum number of prompt tokens specified to have been used over the course of the run.
    - metadata: 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: string
      
      The model that the assistant used for this run.
    - object: "thread.run"
      
      The object type, which is always thread.run.
      - "thread.run"
    - parallel_tool_calls: boolean
      
      Whether to enable parallel function calling during tool use.
    - required_action: object { submit_tool_outputs, type }
      
      Details on the action required to continue the run. Will be null if no action is required.
      - submit_tool_outputs: object { tool_calls }
        
        Details on the tool outputs needed for this run to continue.
        
        tool_calls: array of RequiredActionFunctionToolCall
        
        A list of the relevant tool calls.
        
        id: string
        
        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: object { arguments, name }
        
        The function definition.
        
        arguments: string
        
        The arguments that the model expects you to pass to the function.
        
        name: string
        
        The name of the function.
        
        type: "function"
        
        The type of tool call the output is required for. For now, this is always function.
        
        "function"
      - type: "submit_tool_outputs"
        
        For now, this is always submit_tool_outputs.
        
        "submit_tool_outputs"
    - response_format: 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.
      - "auto"
        
        auto is the default value
        
        "auto"
      - ResponseFormatText object { type }
        
        Default response format. Used to generate text responses.
        
        type: "text"
        
        The type of response format being defined. Always text.
        
        "text"
      - ResponseFormatJSONObject object { type }
        
        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: "json_object"
        
        The type of response format being defined. Always json_object.
        
        "json_object"
      - ResponseFormatJSONSchema object { json_schema, type }
        
        JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
        
        json_schema: object { name, description, schema, strict }
        
        Structured Outputs configuration options, including a JSON Schema.
        
        name: string
        
        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 string
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
        
        schema: optional map[unknown]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
        
        strict: optional boolean
        
        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: "json_schema"
        
        The type of response format being defined. Always json_schema.
        
        "json_schema"
    - started_at: number
      
      The Unix timestamp (in seconds) for when the run was started.
    - status: "queued" or "in_progress" or "requires_action" or 6 more
      
      The status of the run, which can be either queued, in_progress, requires_action, cancelling, cancelled, failed, completed, incomplete, or expired.
      - "queued"
      - "in_progress"
      - "requires_action"
      - "cancelling"
      - "cancelled"
      - "failed"
      - "completed"
      - "incomplete"
      - "expired"
    - thread_id: string
      
      The ID of the thread that was executed on as a part of this run.
    - tool_choice: 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.
      - "none" or "auto" or "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"
      - AssistantToolChoice object { type, function }
        
        Specifies a tool the model should use. Use to force the model to call a specific tool.
        
        type: "function" or "code_interpreter" or "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: string
        
        The name of the function to call.
    - tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool
      
      The list of tools that the assistant used for this run.
      - CodeInterpreterTool object { type }
        
        type: "code_interpreter"
        
        The type of tool being defined: code_interpreter
        
        "code_interpreter"
      - FileSearchTool object { type, file_search }
        
        type: "file_search"
        
        The type of tool being defined: file_search
        
        "file_search"
        
        file_search: optional object { max_num_results, ranking_options }
        
        Overrides for the file search tool.
        
        max_num_results: optional number
        
        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 object { score_threshold, ranker }
        
        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: number
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: optional "auto" or "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"
      - FunctionTool object { function, type }
        
        function: FunctionDefinition
        
        name: string
        
        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 string
        
        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 boolean
        
        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: "function"
        
        The type of tool being defined: function
        
        "function"
    - truncation_strategy: object { type, last_messages }
      
      Controls for how a thread will be truncated prior to the run. Use this to control the initial context window of the run.
      - type: "auto" or "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 number
        
        The number of most recent messages from the thread when constructing the context for the run.
    - usage: object { completion_tokens, prompt_tokens, total_tokens }
      
      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: number
        
        Number of completion tokens used over the course of the run.
      - prompt_tokens: number
        
        Number of prompt tokens used over the course of the run.
      - total_tokens: number
        
        Total number of tokens used (prompt + completion).
    - temperature: optional number
      
      The sampling temperature used for this run. If not set, defaults to 1.
    - top_p: optional number
      
      The nucleus sampling value used for this run. If not set, defaults to 1.
  - event: "thread.run.created"
    - "thread.run.created"
- object { data, event }
  
  Occurs when a run moves to a queued status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.queued"
    - "thread.run.queued"
- object { data, event }
  
  Occurs when a run moves to an in_progress status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.in_progress"
    - "thread.run.in_progress"
- object { data, event }
  
  Occurs when a run moves to a requires_action status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.requires_action"
    - "thread.run.requires_action"
- object { data, event }
  
  Occurs when a run is completed.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.completed"
    - "thread.run.completed"
- object { data, event }
  
  Occurs when a run ends with status incomplete.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.incomplete"
    - "thread.run.incomplete"
- object { data, event }
  
  Occurs when a run fails.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.failed"
    - "thread.run.failed"
- object { data, event }
  
  Occurs when a run moves to a cancelling status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.cancelling"
    - "thread.run.cancelling"
- object { data, event }
  
  Occurs when a run is cancelled.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.cancelled"
    - "thread.run.cancelled"
- object { data, event }
  
  Occurs when a run expires.
  - data: Run
    
    Represents an execution run on a thread.
  - event: "thread.run.expired"
    - "thread.run.expired"

Thread Stream Event

ThreadStreamEvent object { data, event, enabled }

Occurs when a new thread is created.
- data: Thread
  
  Represents a thread that contains messages.
  - id: string
    
    The identifier, which can be referenced in API endpoints.
  - created_at: number
    
    The Unix timestamp (in seconds) for when the thread was created.
  - metadata: 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: "thread"
    
    The object type, which is always thread.
    - "thread"
  - tool_resources: object { code_interpreter, file_search }
    
    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 object { file_ids }
      - file_ids: optional array of string
        
        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 object { vector_store_ids }
      - vector_store_ids: optional array of string
        
        The vector store attached to this thread. There can be a maximum of 1 vector store attached to the thread.
- event: "thread.created"
  - "thread.created"
- enabled: optional boolean
  
  Whether to enable input audio transcription.

resources/beta/subresources/assistants/index.md +326 −466

4 4

5**get** `/assistants`5**get** `/assistants`

6 6

~~7List assistants~~7Returns a list of assistants.

8 8

9### Query Parameters9### Query Parameters

10 10

59 59

60 - `model: string`60 - `model: string`

61 61

62 ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models) for descriptions of them.62 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

63 63

64 - `name: string`64 - `name: string`

65 65

71 71

72 - `"assistant"`72 - `"assistant"`

73 73

~~74 - `tools: array of AssistantTool`~~74 - `tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool`

75 75

76 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`.76 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`.

77 77

99 99

100 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.100 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.

101 101

102 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.102 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

103 103

104 - `ranking_options: optional object { score_threshold, ranker }`104 - `ranking_options: optional object { score_threshold, ranker }`

105 105

106 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.106 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.

107 107

108 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.108 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

109 109

110 - `score_threshold: number`110 - `score_threshold: number`

111 111

133 133

134 - `parameters: optional FunctionParameters`134 - `parameters: optional FunctionParameters`

135 135

136 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.136 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

137 137

138 Omitting `parameters` defines a function with an empty parameter list.138 Omitting `parameters` defines a function with an empty parameter list.

139 139

140 - `strict: optional boolean`140 - `strict: optional boolean`

141 141

142 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](https://platform.openai.com/docs/guides/function-calling).142 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](/docs/guides/function-calling).

143 143

144 - `type: "function"`144 - `type: "function"`

145 145

149 149

150 - `response_format: optional AssistantResponseFormatOption`150 - `response_format: optional AssistantResponseFormatOption`

151 151

152 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.152 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

153 153

154 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](https://platform.openai.com/docs/guides/structured-outputs).154 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](/docs/guides/structured-outputs).

155 155

156 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.156 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

157 157

189 - `ResponseFormatJSONSchema object { json_schema, type }`189 - `ResponseFormatJSONSchema object { json_schema, type }`

190 190

191 JSON Schema response format. Used to generate structured JSON responses.191 JSON Schema response format. Used to generate structured JSON responses.

192 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).192 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

193 193

194 - `json_schema: object { name, description, schema, strict }`194 - `json_schema: object { name, description, schema, strict }`

195 195

216 If set to true, the model will always follow the exact schema defined216 If set to true, the model will always follow the exact schema defined

217 in the `schema` field. Only a subset of JSON Schema is supported when217 in the `schema` field. Only a subset of JSON Schema is supported when

218 `strict` is `true`. To learn more, read the [Structured Outputs218 `strict` is `true`. To learn more, read the [Structured Outputs

219 guide](https://platform.openai.com/docs/guides/structured-outputs).219 guide](/docs/guides/structured-outputs).

220 220

221 - `type: "json_schema"`221 - `type: "json_schema"`

222 222

236 236

237 - `file_ids: optional array of string`237 - `file_ids: optional array of string`

238 238

239 A list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.239 A list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

240 240

241 - `file_search: optional object { vector_store_ids }`241 - `file_search: optional object { vector_store_ids }`

242 242

243 - `vector_store_ids: optional array of string`243 - `vector_store_ids: optional array of string`

244 244

245 The ID of the [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.245 The ID of the [vector store](/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

246 246

247 - `top_p: optional number`247 - `top_p: optional number`

248 248

382 382

383**post** `/assistants`383**post** `/assistants`

384 384

385Create assistant385Create an assistant with a model and instructions.

386 386

387### Body Parameters387### Body Parameters

388 388

389- `model: string or ChatModel`389- `model: string or "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 39 more`

390 390

391 ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models) for descriptions of them.391 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

392 392

393 - `string`393 - `string`

394 394

395 - `ChatModel = "gpt-5.6-sol" or "gpt-5.6-terra" or "gpt-5.6-luna" or 78 more`395 - `AssistantSupportedModels = "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 39 more`

~~396~~

397 - `"gpt-5.6-sol"`

~~398~~

399 - `"gpt-5.6-terra"`

~~400~~

401 - `"gpt-5.6-luna"`

~~402~~

403 - `"gpt-5.4"`

~~404~~

405 - `"gpt-5.4-mini"`

~~406~~

407 - `"gpt-5.4-nano"`

~~408~~

409 - `"gpt-5.4-mini-2026-03-17"`

~~410~~

411 - `"gpt-5.4-nano-2026-03-17"`

~~412~~

413 - `"gpt-5.3-chat-latest"`

~~414~~

415 - `"gpt-5.2"`

~~416~~

417 - `"gpt-5.2-2025-12-11"`

~~418~~

419 - `"gpt-5.2-chat-latest"`

~~420~~

421 - `"gpt-5.2-pro"`

~~422~~

423 - `"gpt-5.2-pro-2025-12-11"`

~~424~~

425 - `"gpt-5.1"`

~~426~~

427 - `"gpt-5.1-2025-11-13"`

~~428~~

429 - `"gpt-5.1-codex"`

~~430~~

431 - `"gpt-5.1-mini"`

432 396

433 - `"gpt-5.1-chat-latest"`397 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

434 398

435 - `"gpt-5"`399 - `"gpt-5"`

436 400

444 408

445 - `"gpt-5-nano-2025-08-07"`409 - `"gpt-5-nano-2025-08-07"`

446 410

447 - `"gpt-5-chat-latest"`

~~448~~

449 - `"gpt-4.1"`411 - `"gpt-4.1"`

450 412

451 - `"gpt-4.1-mini"`413 - `"gpt-4.1-mini"`

458 420

459 - `"gpt-4.1-nano-2025-04-14"`421 - `"gpt-4.1-nano-2025-04-14"`

460 422

461 - `"o4-mini"`

~~462~~

463 - `"o4-mini-2025-04-16"`

~~464~~

465 - `"o3"`

~~466~~

467 - `"o3-2025-04-16"`

~~468~~

469 - `"o3-mini"`423 - `"o3-mini"`

470 424

471 - `"o3-mini-2025-01-31"`425 - `"o3-mini-2025-01-31"`

474 428

475 - `"o1-2024-12-17"`429 - `"o1-2024-12-17"`

476 430

477 - `"o1-preview"`

~~478~~

479 - `"o1-preview-2024-09-12"`

~~480~~

481 - `"o1-mini"`

~~482~~

483 - `"o1-mini-2024-09-12"`

~~484~~

485 - `"gpt-4o"`431 - `"gpt-4o"`

486 432

487 - `"gpt-4o-2024-11-20"`433 - `"gpt-4o-2024-11-20"`

490 436

491 - `"gpt-4o-2024-05-13"`437 - `"gpt-4o-2024-05-13"`

492 438

493 - `"gpt-4o-audio-preview"`

~~494~~

495 - `"gpt-4o-audio-preview-2024-10-01"`

~~496~~

497 - `"gpt-4o-audio-preview-2024-12-17"`

~~498~~

499 - `"gpt-4o-audio-preview-2025-06-03"`

~~500~~

501 - `"gpt-4o-mini-audio-preview"`

~~502~~

503 - `"gpt-4o-mini-audio-preview-2024-12-17"`

~~504~~

505 - `"gpt-4o-search-preview"`

~~506~~

507 - `"gpt-4o-mini-search-preview"`

~~508~~

509 - `"gpt-4o-search-preview-2025-03-11"`

~~510~~

511 - `"gpt-4o-mini-search-preview-2025-03-11"`

~~512~~

513 - `"chatgpt-4o-latest"`

~~514~~

515 - `"codex-mini-latest"`

~~516~~

517 - `"gpt-4o-mini"`439 - `"gpt-4o-mini"`

518 440

519 - `"gpt-4o-mini-2024-07-18"`441 - `"gpt-4o-mini-2024-07-18"`

520 442

443 - `"gpt-4.5-preview"`

444

445 - `"gpt-4.5-preview-2025-02-27"`

446

521 - `"gpt-4-turbo"`447 - `"gpt-4-turbo"`

522 448

523 - `"gpt-4-turbo-2024-04-09"`449 - `"gpt-4-turbo-2024-04-09"`

546 472

547 - `"gpt-3.5-turbo-16k"`473 - `"gpt-3.5-turbo-16k"`

548 474

549 - `"gpt-3.5-turbo-0301"`

~~550~~

551 - `"gpt-3.5-turbo-0613"`475 - `"gpt-3.5-turbo-0613"`

552 476

553 - `"gpt-3.5-turbo-1106"`477 - `"gpt-3.5-turbo-1106"`

603 527

604- `response_format: optional AssistantResponseFormatOption`528- `response_format: optional AssistantResponseFormatOption`

605 529

606 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.530 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

607 531

608 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](https://platform.openai.com/docs/guides/structured-outputs).532 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](/docs/guides/structured-outputs).

609 533

610 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.534 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

611 535

643 - `ResponseFormatJSONSchema object { json_schema, type }`567 - `ResponseFormatJSONSchema object { json_schema, type }`

644 568

645 JSON Schema response format. Used to generate structured JSON responses.569 JSON Schema response format. Used to generate structured JSON responses.

646 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).570 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

647 571

648 - `json_schema: object { name, description, schema, strict }`572 - `json_schema: object { name, description, schema, strict }`

649 573

670 If set to true, the model will always follow the exact schema defined594 If set to true, the model will always follow the exact schema defined

671 in the `schema` field. Only a subset of JSON Schema is supported when595 in the `schema` field. Only a subset of JSON Schema is supported when

672 `strict` is `true`. To learn more, read the [Structured Outputs596 `strict` is `true`. To learn more, read the [Structured Outputs

673 guide](https://platform.openai.com/docs/guides/structured-outputs).597 guide](/docs/guides/structured-outputs).

674 598

675 - `type: "json_schema"`599 - `type: "json_schema"`

676 600

690 614

691 - `file_ids: optional array of string`615 - `file_ids: optional array of string`

692 616

693 A list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter` tool. There can be a maximum of 20 files associated with the tool.617 A list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter` tool. There can be a maximum of 20 files associated with the tool.

694 618

695 - `file_search: optional object { vector_store_ids, vector_stores }`619 - `file_search: optional object { vector_store_ids, vector_stores }`

696 620

697 - `vector_store_ids: optional array of string`621 - `vector_store_ids: optional array of string`

698 622

699 The [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.623 The [vector store](/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

700 624

701 - `vector_stores: optional array of object { chunking_strategy, file_ids, metadata }`625 - `vector_stores: optional array of object { chunking_strategy, file_ids, metadata }`

702 626

703 A helper to create a [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) with file_ids and attach it to this assistant. There can be a maximum of 1 vector store attached to the assistant.627 A helper to create a [vector store](/docs/api-reference/vector-stores/object) with file_ids and attach it to this assistant. There can be a maximum of 1 vector store attached to the assistant.

704 628

705 - `chunking_strategy: optional object { type } or object { static, type }`629 - `chunking_strategy: optional object { type } or object { static, type }`

706 630

707 The chunking strategy used to chunk the file(s). If not set, will use the `auto` strategy.631 The chunking strategy used to chunk the file(s). If not set, will use the `auto` strategy.

708 632

709 - `Auto object { type }`633 - `AutoChunkingStrategy object { type }`

710 634

711 The default strategy. This strategy currently uses a `max_chunk_size_tokens` of `800` and `chunk_overlap_tokens` of `400`.635 The default strategy. This strategy currently uses a `max_chunk_size_tokens` of `800` and `chunk_overlap_tokens` of `400`.

712 636

716 640

717 - `"auto"`641 - `"auto"`

718 642

719 - `Static object { static, type }`643 - `StaticChunkingStrategy object { static, type }`

720 644

721 - `static: object { chunk_overlap_tokens, max_chunk_size_tokens }`645 - `static: object { chunk_overlap_tokens, max_chunk_size_tokens }`

722 646

738 662

739 - `file_ids: optional array of string`663 - `file_ids: optional array of string`

740 664

741 A list of [file](https://platform.openai.com/docs/api-reference/files) 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.665 A list of [file](/docs/api-reference/files) 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.

742 666

743 - `metadata: optional Metadata`667 - `metadata: optional Metadata`

744 668

749 Keys are strings with a maximum length of 64 characters. Values are strings673 Keys are strings with a maximum length of 64 characters. Values are strings

750 with a maximum length of 512 characters.674 with a maximum length of 512 characters.

751 675

752- `tools: optional array of AssistantTool`676- `tools: optional array of CodeInterpreterTool or FileSearchTool or FunctionTool`

753 677

754 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`.678 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`.

755 679

777 701

778 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.702 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.

779 703

780 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.704 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

781 705

782 - `ranking_options: optional object { score_threshold, ranker }`706 - `ranking_options: optional object { score_threshold, ranker }`

783 707

784 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.708 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.

785 709

786 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.710 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

787 711

788 - `score_threshold: number`712 - `score_threshold: number`

789 713

811 735

812 - `parameters: optional FunctionParameters`736 - `parameters: optional FunctionParameters`

813 737

814 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.738 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

815 739

816 Omitting `parameters` defines a function with an empty parameter list.740 Omitting `parameters` defines a function with an empty parameter list.

817 741

818 - `strict: optional boolean`742 - `strict: optional boolean`

819 743

820 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](https://platform.openai.com/docs/guides/function-calling).744 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](/docs/guides/function-calling).

821 745

822 - `type: "function"`746 - `type: "function"`

823 747

864 788

865 - `model: string`789 - `model: string`

866 790

867 ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models) for descriptions of them.791 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

868 792

869 - `name: string`793 - `name: string`

870 794

876 800

877 - `"assistant"`801 - `"assistant"`

878 802

879 - `tools: array of AssistantTool`803 - `tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool`

880 804

881 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`.805 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`.

882 806

904 828

905 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.829 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.

906 830

907 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.831 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

908 832

909 - `ranking_options: optional object { score_threshold, ranker }`833 - `ranking_options: optional object { score_threshold, ranker }`

910 834

911 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.835 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.

912 836

913 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.837 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

914 838

915 - `score_threshold: number`839 - `score_threshold: number`

916 840

938 862

939 - `parameters: optional FunctionParameters`863 - `parameters: optional FunctionParameters`

940 864

941 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.865 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

942 866

943 Omitting `parameters` defines a function with an empty parameter list.867 Omitting `parameters` defines a function with an empty parameter list.

944 868

945 - `strict: optional boolean`869 - `strict: optional boolean`

946 870

947 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](https://platform.openai.com/docs/guides/function-calling).871 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](/docs/guides/function-calling).

948 872

949 - `type: "function"`873 - `type: "function"`

950 874

954 878

955 - `response_format: optional AssistantResponseFormatOption`879 - `response_format: optional AssistantResponseFormatOption`

956 880

957 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.881 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

958 882

959 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](https://platform.openai.com/docs/guides/structured-outputs).883 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](/docs/guides/structured-outputs).

960 884

961 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.885 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

962 886

994 - `ResponseFormatJSONSchema object { json_schema, type }`918 - `ResponseFormatJSONSchema object { json_schema, type }`

995 919

996 JSON Schema response format. Used to generate structured JSON responses.920 JSON Schema response format. Used to generate structured JSON responses.

997 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).921 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

998 922

999 - `json_schema: object { name, description, schema, strict }`923 - `json_schema: object { name, description, schema, strict }`

1000 924

1021 If set to true, the model will always follow the exact schema defined945 If set to true, the model will always follow the exact schema defined

1022 in the `schema` field. Only a subset of JSON Schema is supported when946 in the `schema` field. Only a subset of JSON Schema is supported when

1023 `strict` is `true`. To learn more, read the [Structured Outputs947 `strict` is `true`. To learn more, read the [Structured Outputs

1024 guide](https://platform.openai.com/docs/guides/structured-outputs).948 guide](/docs/guides/structured-outputs).

1025 949

1026 - `type: "json_schema"`950 - `type: "json_schema"`

1027 951

1041 965

1042 - `file_ids: optional array of string`966 - `file_ids: optional array of string`

1043 967

1044 A list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.968 A list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

1045 969

1046 - `file_search: optional object { vector_store_ids }`970 - `file_search: optional object { vector_store_ids }`

1047 971

1048 - `vector_store_ids: optional array of string`972 - `vector_store_ids: optional array of string`

1049 973

1050 The ID of the [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.974 The ID of the [vector store](/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

1051 975

1052 - `top_p: optional number`976 - `top_p: optional number`

1053 977

1191 1115

1192**get** `/assistants/{assistant_id}`1116**get** `/assistants/{assistant_id}`

1193 1117

1194Retrieve assistant1118Retrieves an assistant.

1195 1119

1196### Path Parameters1120### Path Parameters

1197 1121

1230 1154

1231 - `model: string`1155 - `model: string`

1232 1156

1233 ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models) for descriptions of them.1157 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

1234 1158

1235 - `name: string`1159 - `name: string`

1236 1160

1242 1166

1243 - `"assistant"`1167 - `"assistant"`

1244 1168

1245 - `tools: array of AssistantTool`1169 - `tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool`

1246 1170

1247 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`.1171 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`.

1248 1172

1270 1194

1271 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.1195 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.

1272 1196

1273 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.1197 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1274 1198

1275 - `ranking_options: optional object { score_threshold, ranker }`1199 - `ranking_options: optional object { score_threshold, ranker }`

1276 1200

1277 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.1201 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.

1278 1202

1279 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.1203 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1280 1204

1281 - `score_threshold: number`1205 - `score_threshold: number`

1282 1206

1304 1228

1305 - `parameters: optional FunctionParameters`1229 - `parameters: optional FunctionParameters`

1306 1230

1307 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.1231 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

1308 1232

1309 Omitting `parameters` defines a function with an empty parameter list.1233 Omitting `parameters` defines a function with an empty parameter list.

1310 1234

1311 - `strict: optional boolean`1235 - `strict: optional boolean`

1312 1236

1313 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](https://platform.openai.com/docs/guides/function-calling).1237 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](/docs/guides/function-calling).

1314 1238

1315 - `type: "function"`1239 - `type: "function"`

1316 1240

1320 1244

1321 - `response_format: optional AssistantResponseFormatOption`1245 - `response_format: optional AssistantResponseFormatOption`

1322 1246

1323 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.1247 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

1324 1248

1325 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](https://platform.openai.com/docs/guides/structured-outputs).1249 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](/docs/guides/structured-outputs).

1326 1250

1327 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.1251 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

1328 1252

1360 - `ResponseFormatJSONSchema object { json_schema, type }`1284 - `ResponseFormatJSONSchema object { json_schema, type }`

1361 1285

1362 JSON Schema response format. Used to generate structured JSON responses.1286 JSON Schema response format. Used to generate structured JSON responses.

1363 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).1287 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

1364 1288

1365 - `json_schema: object { name, description, schema, strict }`1289 - `json_schema: object { name, description, schema, strict }`

1366 1290

1387 If set to true, the model will always follow the exact schema defined1311 If set to true, the model will always follow the exact schema defined

1388 in the `schema` field. Only a subset of JSON Schema is supported when1312 in the `schema` field. Only a subset of JSON Schema is supported when

1389 `strict` is `true`. To learn more, read the [Structured Outputs1313 `strict` is `true`. To learn more, read the [Structured Outputs

1390 guide](https://platform.openai.com/docs/guides/structured-outputs).1314 guide](/docs/guides/structured-outputs).

1391 1315

1392 - `type: "json_schema"`1316 - `type: "json_schema"`

1393 1317

1407 1331

1408 - `file_ids: optional array of string`1332 - `file_ids: optional array of string`

1409 1333

1410 A list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.1334 A list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

1411 1335

1412 - `file_search: optional object { vector_store_ids }`1336 - `file_search: optional object { vector_store_ids }`

1413 1337

1414 - `vector_store_ids: optional array of string`1338 - `vector_store_ids: optional array of string`

1415 1339

1416 The ID of the [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.1340 The ID of the [vector store](/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

1417 1341

1418 - `top_p: optional number`1342 - `top_p: optional number`

1419 1343

1502 1426

1503**post** `/assistants/{assistant_id}`1427**post** `/assistants/{assistant_id}`

1504 1428

1505Modify assistant1429Modifies an assistant.

1506 1430

1507### Path Parameters1431### Path Parameters

1508 1432

1529 1453

1530- `model: optional string or "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 39 more`1454- `model: optional string or "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 39 more`

1531 1455

1532 ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models) for descriptions of them.1456 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

1533 1457

1534 - `string`1458 - `string`

1535 1459

1536 - `AssistantSupportedModels = "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 39 more`1460 - `AssistantSupportedModels = "gpt-5" or "gpt-5-mini" or "gpt-5-nano" or 39 more`

1537 1461

1538 ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models) for descriptions of them.1462 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

1539 1463

1540 - `"gpt-5"`1464 - `"gpt-5"`

1541 1465

1651 1575

1652- `response_format: optional AssistantResponseFormatOption`1576- `response_format: optional AssistantResponseFormatOption`

1653 1577

1654 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.1578 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

1655 1579

1656 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](https://platform.openai.com/docs/guides/structured-outputs).1580 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](/docs/guides/structured-outputs).

1657 1581

1658 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.1582 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

1659 1583

1691 - `ResponseFormatJSONSchema object { json_schema, type }`1615 - `ResponseFormatJSONSchema object { json_schema, type }`

1692 1616

1693 JSON Schema response format. Used to generate structured JSON responses.1617 JSON Schema response format. Used to generate structured JSON responses.

1694 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).1618 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

1695 1619

1696 - `json_schema: object { name, description, schema, strict }`1620 - `json_schema: object { name, description, schema, strict }`

1697 1621

1718 If set to true, the model will always follow the exact schema defined1642 If set to true, the model will always follow the exact schema defined

1719 in the `schema` field. Only a subset of JSON Schema is supported when1643 in the `schema` field. Only a subset of JSON Schema is supported when

1720 `strict` is `true`. To learn more, read the [Structured Outputs1644 `strict` is `true`. To learn more, read the [Structured Outputs

1721 guide](https://platform.openai.com/docs/guides/structured-outputs).1645 guide](/docs/guides/structured-outputs).

1722 1646

1723 - `type: "json_schema"`1647 - `type: "json_schema"`

1724 1648

1738 1662

1739 - `file_ids: optional array of string`1663 - `file_ids: optional array of string`

1740 1664

1741 Overrides the list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter` tool. There can be a maximum of 20 files associated with the tool.1665 Overrides the list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter` tool. There can be a maximum of 20 files associated with the tool.

1742 1666

1743 - `file_search: optional object { vector_store_ids }`1667 - `file_search: optional object { vector_store_ids }`

1744 1668

1745 - `vector_store_ids: optional array of string`1669 - `vector_store_ids: optional array of string`

1746 1670

1747 Overrides the [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.1671 Overrides the [vector store](/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

1748 1672

1749- `tools: optional array of AssistantTool`1673- `tools: optional array of CodeInterpreterTool or FileSearchTool or FunctionTool`

1750 1674

1751 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`.1675 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`.

1752 1676

1774 1698

1775 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.1699 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.

1776 1700

1777 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.1701 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1778 1702

1779 - `ranking_options: optional object { score_threshold, ranker }`1703 - `ranking_options: optional object { score_threshold, ranker }`

1780 1704

1781 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.1705 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.

1782 1706

1783 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.1707 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1784 1708

1785 - `score_threshold: number`1709 - `score_threshold: number`

1786 1710

1808 1732

1809 - `parameters: optional FunctionParameters`1733 - `parameters: optional FunctionParameters`

1810 1734

1811 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.1735 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

1812 1736

1813 Omitting `parameters` defines a function with an empty parameter list.1737 Omitting `parameters` defines a function with an empty parameter list.

1814 1738

1815 - `strict: optional boolean`1739 - `strict: optional boolean`

1816 1740

1817 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](https://platform.openai.com/docs/guides/function-calling).1741 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](/docs/guides/function-calling).

1818 1742

1819 - `type: "function"`1743 - `type: "function"`

1820 1744

1861 1785

1862 - `model: string`1786 - `model: string`

1863 1787

1864 ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models) for descriptions of them.1788 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

1865 1789

1866 - `name: string`1790 - `name: string`

1867 1791

1873 1797

1874 - `"assistant"`1798 - `"assistant"`

1875 1799

1876 - `tools: array of AssistantTool`1800 - `tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool`

1877 1801

1878 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`.1802 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`.

1879 1803

1901 1825

1902 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.1826 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.

1903 1827

1904 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.1828 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1905 1829

1906 - `ranking_options: optional object { score_threshold, ranker }`1830 - `ranking_options: optional object { score_threshold, ranker }`

1907 1831

1908 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.1832 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.

1909 1833

1910 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.1834 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1911 1835

1912 - `score_threshold: number`1836 - `score_threshold: number`

1913 1837

1935 1859

1936 - `parameters: optional FunctionParameters`1860 - `parameters: optional FunctionParameters`

1937 1861

1938 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.1862 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

1939 1863

1940 Omitting `parameters` defines a function with an empty parameter list.1864 Omitting `parameters` defines a function with an empty parameter list.

1941 1865

1942 - `strict: optional boolean`1866 - `strict: optional boolean`

1943 1867

1944 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](https://platform.openai.com/docs/guides/function-calling).1868 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](/docs/guides/function-calling).

1945 1869

1946 - `type: "function"`1870 - `type: "function"`

1947 1871

1951 1875

1952 - `response_format: optional AssistantResponseFormatOption`1876 - `response_format: optional AssistantResponseFormatOption`

1953 1877

1954 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.1878 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

1955 1879

1956 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](https://platform.openai.com/docs/guides/structured-outputs).1880 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](/docs/guides/structured-outputs).

1957 1881

1958 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.1882 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

1959 1883

1991 - `ResponseFormatJSONSchema object { json_schema, type }`1915 - `ResponseFormatJSONSchema object { json_schema, type }`

1992 1916

1993 JSON Schema response format. Used to generate structured JSON responses.1917 JSON Schema response format. Used to generate structured JSON responses.

1994 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).1918 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

1995 1919

1996 - `json_schema: object { name, description, schema, strict }`1920 - `json_schema: object { name, description, schema, strict }`

1997 1921

2018 If set to true, the model will always follow the exact schema defined1942 If set to true, the model will always follow the exact schema defined

2019 in the `schema` field. Only a subset of JSON Schema is supported when1943 in the `schema` field. Only a subset of JSON Schema is supported when

2020 `strict` is `true`. To learn more, read the [Structured Outputs1944 `strict` is `true`. To learn more, read the [Structured Outputs

2021 guide](https://platform.openai.com/docs/guides/structured-outputs).1945 guide](/docs/guides/structured-outputs).

2022 1946

2023 - `type: "json_schema"`1947 - `type: "json_schema"`

2024 1948

2038 1962

2039 - `file_ids: optional array of string`1963 - `file_ids: optional array of string`

2040 1964

2041 A list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.1965 A list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

2042 1966

2043 - `file_search: optional object { vector_store_ids }`1967 - `file_search: optional object { vector_store_ids }`

2044 1968

2045 - `vector_store_ids: optional array of string`1969 - `vector_store_ids: optional array of string`

2046 1970

2047 The ID of the [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.1971 The ID of the [vector store](/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

2048 1972

2049 - `top_p: optional number`1973 - `top_p: optional number`

2050 1974

2148 2072

2149**delete** `/assistants/{assistant_id}`2073**delete** `/assistants/{assistant_id}`

2150 2074

2151Delete assistant2075Delete an assistant.

2152 2076

2153### Path Parameters2077### Path Parameters

2154 2078

2240 2164

2241 - `model: string`2165 - `model: string`

2242 2166

2243 ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models) for descriptions of them.2167 ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models) for descriptions of them.

2244 2168

2245 - `name: string`2169 - `name: string`

2246 2170

2252 2176

2253 - `"assistant"`2177 - `"assistant"`

2254 2178

2255 - `tools: array of AssistantTool`2179 - `tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool`

2256 2180

2257 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`.2181 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`.

2258 2182

2280 2204

2281 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.2205 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.

2282 2206

2283 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.2207 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

2284 2208

2285 - `ranking_options: optional object { score_threshold, ranker }`2209 - `ranking_options: optional object { score_threshold, ranker }`

2286 2210

2287 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.2211 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.

2288 2212

2289 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.2213 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

2290 2214

2291 - `score_threshold: number`2215 - `score_threshold: number`

2292 2216

2314 2238

2315 - `parameters: optional FunctionParameters`2239 - `parameters: optional FunctionParameters`

2316 2240

2317 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.2241 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

2318 2242

2319 Omitting `parameters` defines a function with an empty parameter list.2243 Omitting `parameters` defines a function with an empty parameter list.

2320 2244

2321 - `strict: optional boolean`2245 - `strict: optional boolean`

2322 2246

2323 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](https://platform.openai.com/docs/guides/function-calling).2247 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](/docs/guides/function-calling).

2324 2248

2325 - `type: "function"`2249 - `type: "function"`

2326 2250

2330 2254

2331 - `response_format: optional AssistantResponseFormatOption`2255 - `response_format: optional AssistantResponseFormatOption`

2332 2256

2333 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.2257 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

2334 2258

2335 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](https://platform.openai.com/docs/guides/structured-outputs).2259 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](/docs/guides/structured-outputs).

2336 2260

2337 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.2261 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

2338 2262

2370 - `ResponseFormatJSONSchema object { json_schema, type }`2294 - `ResponseFormatJSONSchema object { json_schema, type }`

2371 2295

2372 JSON Schema response format. Used to generate structured JSON responses.2296 JSON Schema response format. Used to generate structured JSON responses.

2373 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).2297 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

2374 2298

2375 - `json_schema: object { name, description, schema, strict }`2299 - `json_schema: object { name, description, schema, strict }`

2376 2300

2397 If set to true, the model will always follow the exact schema defined2321 If set to true, the model will always follow the exact schema defined

2398 in the `schema` field. Only a subset of JSON Schema is supported when2322 in the `schema` field. Only a subset of JSON Schema is supported when

2399 `strict` is `true`. To learn more, read the [Structured Outputs2323 `strict` is `true`. To learn more, read the [Structured Outputs

2400 guide](https://platform.openai.com/docs/guides/structured-outputs).2324 guide](/docs/guides/structured-outputs).

2401 2325

2402 - `type: "json_schema"`2326 - `type: "json_schema"`

2403 2327

2417 2341

2418 - `file_ids: optional array of string`2342 - `file_ids: optional array of string`

2419 2343

2420 A list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.2344 A list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter`` tool. There can be a maximum of 20 files associated with the tool.

2421 2345

2422 - `file_search: optional object { vector_store_ids }`2346 - `file_search: optional object { vector_store_ids }`

2423 2347

2424 - `vector_store_ids: optional array of string`2348 - `vector_store_ids: optional array of string`

2425 2349

2426 The ID of the [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.2350 The ID of the [vector store](/docs/api-reference/vector-stores/object) attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.

2427 2351

2428 - `top_p: optional number`2352 - `top_p: optional number`

2429 2353

2445 2369

2446### Assistant Stream Event2370### Assistant Stream Event

2447 2371

2448- `AssistantStreamEvent = object { data, event, enabled } or object { data, event } or object { data, event } or 21 more`2372- `AssistantStreamEvent = object { data, event, enabled } or object { data, event } or object { data, event } or 22 more`

2449 2373

2450 Represents an event emitted when streaming a Run.2374 Represents an event emitted when streaming a Run.

2451 2375

2464 `thread.message.completed` event.2388 `thread.message.completed` event.

2465 2389

2466 We may add additional events over time, so we recommend handling unknown events gracefully2390 We may add additional events over time, so we recommend handling unknown events gracefully

2467 in your code. See the [Assistants API quickstart](https://platform.openai.com/docs/assistants/overview) to learn how to2391 in your code. See the [Assistants API quickstart](/docs/assistants/overview) to learn how to

2468 integrate the Assistants API with streaming.2392 integrate the Assistants API with streaming.

2469 2393

2470 - `ThreadCreated object { data, event, enabled }`2394 - `object { data, event, enabled }`

2471 2395

2472 Occurs when a new [thread](https://platform.openai.com/docs/api-reference/threads/object) is created.2396 Occurs when a new [thread](/docs/api-reference/threads/object) is created.

2473 2397

2474 - `data: Thread`2398 - `data: Thread`

2475 2399

2476 Represents a thread that contains [messages](https://platform.openai.com/docs/api-reference/messages).2400 Represents a thread that contains [messages](/docs/api-reference/messages).

2477 2401

2478 - `id: string`2402 - `id: string`

2479 2403

2506 2430

2507 - `file_ids: optional array of string`2431 - `file_ids: optional array of string`

2508 2432

2509 A list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter` tool. There can be a maximum of 20 files associated with the tool.2433 A list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter` tool. There can be a maximum of 20 files associated with the tool.

2510 2434

2511 - `file_search: optional object { vector_store_ids }`2435 - `file_search: optional object { vector_store_ids }`

2512 2436

2513 - `vector_store_ids: optional array of string`2437 - `vector_store_ids: optional array of string`

2514 2438

2515 The [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this thread. There can be a maximum of 1 vector store attached to the thread.2439 The [vector store](/docs/api-reference/vector-stores/object) attached to this thread. There can be a maximum of 1 vector store attached to the thread.

2516 2440

2517 - `event: "thread.created"`2441 - `event: "thread.created"`

2518 2442

2522 2446

2523 Whether to enable input audio transcription.2447 Whether to enable input audio transcription.

2524 2448

2525 - `ThreadRunCreated object { data, event }`2449 - `object { data, event }`

2526 2450

2527 Occurs when a new [run](https://platform.openai.com/docs/api-reference/runs/object) is created.2451 Occurs when a new [run](/docs/api-reference/runs/object) is created.

2528 2452

2529 - `data: Run`2453 - `data: Run`

2530 2454

2531 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2455 Represents an execution run on a [thread](/docs/api-reference/threads).

2532 2456

2533 - `id: string`2457 - `id: string`

2534 2458

2536 2460

2537 - `assistant_id: string`2461 - `assistant_id: string`

2538 2462

2539 The ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for execution of this run.2463 The ID of the [assistant](/docs/api-reference/assistants) used for execution of this run.

2540 2464

2541 - `cancelled_at: number`2465 - `cancelled_at: number`

2542 2466

2572 2496

2573 - `instructions: string`2497 - `instructions: string`

2574 2498

2575 The instructions that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.2499 The instructions that the [assistant](/docs/api-reference/assistants) used for this run.

2576 2500

2577 - `last_error: object { code, message }`2501 - `last_error: object { code, message }`

2578 2502

2611 2535

2612 - `model: string`2536 - `model: string`

2613 2537

2614 The model that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.2538 The model that the [assistant](/docs/api-reference/assistants) used for this run.

2615 2539

2616 - `object: "thread.run"`2540 - `object: "thread.run"`

2617 2541

2621 2545

2622 - `parallel_tool_calls: boolean`2546 - `parallel_tool_calls: boolean`

2623 2547

2624 Whether to enable [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling) during tool use.2548 Whether to enable [parallel function calling](/docs/guides/function-calling#configuring-parallel-function-calling) during tool use.

2625 2549

2626 - `required_action: object { submit_tool_outputs, type }`2550 - `required_action: object { submit_tool_outputs, type }`

2627 2551

2637 2561

2638 - `id: string`2562 - `id: string`

2639 2563

2640 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](https://platform.openai.com/docs/api-reference/runs/submitToolOutputs) endpoint.2564 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](/docs/api-reference/runs/submitToolOutputs) endpoint.

2641 2565

2642 - `function: object { arguments, name }`2566 - `function: object { arguments, name }`

2643 2567

2665 2589

2666 - `response_format: AssistantResponseFormatOption`2590 - `response_format: AssistantResponseFormatOption`

2667 2591

2668 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.2592 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

2669 2593

2670 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](https://platform.openai.com/docs/guides/structured-outputs).2594 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](/docs/guides/structured-outputs).

2671 2595

2672 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.2596 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

2673 2597

2705 - `ResponseFormatJSONSchema object { json_schema, type }`2629 - `ResponseFormatJSONSchema object { json_schema, type }`

2706 2630

2707 JSON Schema response format. Used to generate structured JSON responses.2631 JSON Schema response format. Used to generate structured JSON responses.

2708 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).2632 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

2709 2633

2710 - `json_schema: object { name, description, schema, strict }`2634 - `json_schema: object { name, description, schema, strict }`

2711 2635

2732 If set to true, the model will always follow the exact schema defined2656 If set to true, the model will always follow the exact schema defined

2733 in the `schema` field. Only a subset of JSON Schema is supported when2657 in the `schema` field. Only a subset of JSON Schema is supported when

2734 `strict` is `true`. To learn more, read the [Structured Outputs2658 `strict` is `true`. To learn more, read the [Structured Outputs

2735 guide](https://platform.openai.com/docs/guides/structured-outputs).2659 guide](/docs/guides/structured-outputs).

2736 2660

2737 - `type: "json_schema"`2661 - `type: "json_schema"`

2738 2662

2744 2668

2745 The Unix timestamp (in seconds) for when the run was started.2669 The Unix timestamp (in seconds) for when the run was started.

2746 2670

2747 - `status: RunStatus`2671 - `status: "queued" or "in_progress" or "requires_action" or 6 more`

2748 2672

2749 The status of the run, which can be either `queued`, `in_progress`, `requires_action`, `cancelling`, `cancelled`, `failed`, `completed`, `incomplete`, or `expired`.2673 The status of the run, which can be either `queued`, `in_progress`, `requires_action`, `cancelling`, `cancelled`, `failed`, `completed`, `incomplete`, or `expired`.

2750 2674

2768 2692

2769 - `thread_id: string`2693 - `thread_id: string`

2770 2694

2771 The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) that was executed on as a part of this run.2695 The ID of the [thread](/docs/api-reference/threads) that was executed on as a part of this run.

2772 2696

2773 - `tool_choice: AssistantToolChoiceOption`2697 - `tool_choice: AssistantToolChoiceOption`

2774 2698

2778 `required` means the model must call one or more tools before responding to the user.2702 `required` means the model must call one or more tools before responding to the user.

2779 Specifying a particular tool like `{"type": "file_search"}` or `{"type": "function", "function": {"name": "my_function"}}` forces the model to call that tool.2703 Specifying a particular tool like `{"type": "file_search"}` or `{"type": "function", "function": {"name": "my_function"}}` forces the model to call that tool.

2780 2704

2781 - `Auto = "none" or "auto" or "required"`2705 - `"none" or "auto" or "required"`

2782 2706

2783 `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.2707 `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.

2784 2708

2808 2732

2809 The name of the function to call.2733 The name of the function to call.

2810 2734

2811 - `tools: array of AssistantTool`2735 - `tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool`

2812 2736

2813 The list of tools that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.2737 The list of tools that the [assistant](/docs/api-reference/assistants) used for this run.

2814 2738

2815 - `CodeInterpreterTool object { type }`2739 - `CodeInterpreterTool object { type }`

2816 2740

2836 2760

2837 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.2761 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.

2838 2762

2839 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.2763 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

2840 2764

2841 - `ranking_options: optional object { score_threshold, ranker }`2765 - `ranking_options: optional object { score_threshold, ranker }`

2842 2766

2843 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.2767 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.

2844 2768

2845 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.2769 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

2846 2770

2847 - `score_threshold: number`2771 - `score_threshold: number`

2848 2772

2870 2794

2871 - `parameters: optional FunctionParameters`2795 - `parameters: optional FunctionParameters`

2872 2796

2873 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.2797 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

2874 2798

2875 Omitting `parameters` defines a function with an empty parameter list.2799 Omitting `parameters` defines a function with an empty parameter list.

2876 2800

2877 - `strict: optional boolean`2801 - `strict: optional boolean`

2878 2802

2879 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](https://platform.openai.com/docs/guides/function-calling).2803 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](/docs/guides/function-calling).

2880 2804

2881 - `type: "function"`2805 - `type: "function"`

2882 2806

2928 2852

2929 - `"thread.run.created"`2853 - `"thread.run.created"`

2930 2854

2931 - `ThreadRunQueued object { data, event }`2855 - `object { data, event }`

2932 2856

2933 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) moves to a `queued` status.2857 Occurs when a [run](/docs/api-reference/runs/object) moves to a `queued` status.

2934 2858

2935 - `data: Run`2859 - `data: Run`

2936 2860

2937 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2861 Represents an execution run on a [thread](/docs/api-reference/threads).

2938 2862

2939 - `event: "thread.run.queued"`2863 - `event: "thread.run.queued"`

2940 2864

2941 - `"thread.run.queued"`2865 - `"thread.run.queued"`

2942 2866

2943 - `ThreadRunInProgress object { data, event }`2867 - `object { data, event }`

2944 2868

2945 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) moves to an `in_progress` status.2869 Occurs when a [run](/docs/api-reference/runs/object) moves to an `in_progress` status.

2946 2870

2947 - `data: Run`2871 - `data: Run`

2948 2872

2949 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2873 Represents an execution run on a [thread](/docs/api-reference/threads).

2950 2874

2951 - `event: "thread.run.in_progress"`2875 - `event: "thread.run.in_progress"`

2952 2876

2953 - `"thread.run.in_progress"`2877 - `"thread.run.in_progress"`

2954 2878

2955 - `ThreadRunRequiresAction object { data, event }`2879 - `object { data, event }`

2956 2880

2957 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) moves to a `requires_action` status.2881 Occurs when a [run](/docs/api-reference/runs/object) moves to a `requires_action` status.

2958 2882

2959 - `data: Run`2883 - `data: Run`

2960 2884

2961 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2885 Represents an execution run on a [thread](/docs/api-reference/threads).

2962 2886

2963 - `event: "thread.run.requires_action"`2887 - `event: "thread.run.requires_action"`

2964 2888

2965 - `"thread.run.requires_action"`2889 - `"thread.run.requires_action"`

2966 2890

2967 - `ThreadRunCompleted object { data, event }`2891 - `object { data, event }`

2968 2892

2969 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) is completed.2893 Occurs when a [run](/docs/api-reference/runs/object) is completed.

2970 2894

2971 - `data: Run`2895 - `data: Run`

2972 2896

2973 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2897 Represents an execution run on a [thread](/docs/api-reference/threads).

2974 2898

2975 - `event: "thread.run.completed"`2899 - `event: "thread.run.completed"`

2976 2900

2977 - `"thread.run.completed"`2901 - `"thread.run.completed"`

2978 2902

2979 - `ThreadRunIncomplete object { data, event }`2903 - `object { data, event }`

2980 2904

2981 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) ends with status `incomplete`.2905 Occurs when a [run](/docs/api-reference/runs/object) ends with status `incomplete`.

2982 2906

2983 - `data: Run`2907 - `data: Run`

2984 2908

2985 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2909 Represents an execution run on a [thread](/docs/api-reference/threads).

2986 2910

2987 - `event: "thread.run.incomplete"`2911 - `event: "thread.run.incomplete"`

2988 2912

2989 - `"thread.run.incomplete"`2913 - `"thread.run.incomplete"`

2990 2914

2991 - `ThreadRunFailed object { data, event }`2915 - `object { data, event }`

2992 2916

2993 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) fails.2917 Occurs when a [run](/docs/api-reference/runs/object) fails.

2994 2918

2995 - `data: Run`2919 - `data: Run`

2996 2920

2997 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2921 Represents an execution run on a [thread](/docs/api-reference/threads).

2998 2922

2999 - `event: "thread.run.failed"`2923 - `event: "thread.run.failed"`

3000 2924

3001 - `"thread.run.failed"`2925 - `"thread.run.failed"`

3002 2926

3003 - `ThreadRunCancelling object { data, event }`2927 - `object { data, event }`

3004 2928

3005 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) moves to a `cancelling` status.2929 Occurs when a [run](/docs/api-reference/runs/object) moves to a `cancelling` status.

3006 2930

3007 - `data: Run`2931 - `data: Run`

3008 2932

3009 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2933 Represents an execution run on a [thread](/docs/api-reference/threads).

3010 2934

3011 - `event: "thread.run.cancelling"`2935 - `event: "thread.run.cancelling"`

3012 2936

3013 - `"thread.run.cancelling"`2937 - `"thread.run.cancelling"`

3014 2938

3015 - `ThreadRunCancelled object { data, event }`2939 - `object { data, event }`

3016 2940

3017 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) is cancelled.2941 Occurs when a [run](/docs/api-reference/runs/object) is cancelled.

3018 2942

3019 - `data: Run`2943 - `data: Run`

3020 2944

3021 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2945 Represents an execution run on a [thread](/docs/api-reference/threads).

3022 2946

3023 - `event: "thread.run.cancelled"`2947 - `event: "thread.run.cancelled"`

3024 2948

3025 - `"thread.run.cancelled"`2949 - `"thread.run.cancelled"`

3026 2950

3027 - `ThreadRunExpired object { data, event }`2951 - `object { data, event }`

3028 2952

3029 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) expires.2953 Occurs when a [run](/docs/api-reference/runs/object) expires.

3030 2954

3031 - `data: Run`2955 - `data: Run`

3032 2956

3033 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).2957 Represents an execution run on a [thread](/docs/api-reference/threads).

3034 2958

3035 - `event: "thread.run.expired"`2959 - `event: "thread.run.expired"`

3036 2960

3037 - `"thread.run.expired"`2961 - `"thread.run.expired"`

3038 2962

3039 - `ThreadRunStepCreated object { data, event }`2963 - `object { data, event }`

3040 2964

3041 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is created.2965 Occurs when a [run step](/docs/api-reference/run-steps/step-object) is created.

3042 2966

3043 - `data: RunStep`2967 - `data: RunStep`

3044 2968

3050 2974

3051 - `assistant_id: string`2975 - `assistant_id: string`

3052 2976

3053 The ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) associated with the run step.2977 The ID of the [assistant](/docs/api-reference/assistants) associated with the run step.

3054 2978

3055 - `cancelled_at: number`2979 - `cancelled_at: number`

3056 2980

3105 3029

3106 - `run_id: string`3030 - `run_id: string`

3107 3031

3108 The ID of the [run](https://platform.openai.com/docs/api-reference/runs) that this run step is a part of.3032 The ID of the [run](/docs/api-reference/runs) that this run step is a part of.

3109 3033

3110 - `status: "in_progress" or "cancelled" or "failed" or 2 more`3034 - `status: "in_progress" or "cancelled" or "failed" or 2 more`

3111 3035

3145 3069

3146 Details of the tool call.3070 Details of the tool call.

3147 3071

3148 - `tool_calls: array of ToolCall`3072 - `tool_calls: array of CodeInterpreterToolCall or FileSearchToolCall or FunctionToolCall`

3149 3073

3150 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`.3074 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`.

3151 3075

3169 3093

3170 The outputs from the Code Interpreter tool call. Code Interpreter can output one or more items, including text (`logs`) or images (`image`). Each of these are represented by a different object type.3094 The outputs from the Code Interpreter tool call. Code Interpreter can output one or more items, including text (`logs`) or images (`image`). Each of these are represented by a different object type.

3171 3095

3172 - `Logs object { logs, type }`3096 - `CodeInterpreterLogOutput object { logs, type }`

3173 3097

3174 Text output from the Code Interpreter tool call as part of a run step.3098 Text output from the Code Interpreter tool call as part of a run step.

3175 3099

3183 3107

3184 - `"logs"`3108 - `"logs"`

3185 3109

3186 - `Image object { image, type }`3110 - `CodeInterpreterImageOutput object { image, type }`

3187 3111

3188 - `image: object { file_id }`3112 - `image: object { file_id }`

3189 3113

3190 - `file_id: string`3114 - `file_id: string`

3191 3115

3192 The [file](https://platform.openai.com/docs/api-reference/files) ID of the image.3116 The [file](/docs/api-reference/files) ID of the image.

3193 3117

3194 - `type: "image"`3118 - `type: "image"`

3195 3119

3285 3209

3286 - `output: string`3210 - `output: string`

3287 3211

3288 The output of the function. This will be `null` if the outputs have not been [submitted](https://platform.openai.com/docs/api-reference/runs/submitToolOutputs) yet.3212 The output of the function. This will be `null` if the outputs have not been [submitted](/docs/api-reference/runs/submitToolOutputs) yet.

3289 3213

3290 - `type: "function"`3214 - `type: "function"`

3291 3215

3301 3225

3302 - `thread_id: string`3226 - `thread_id: string`

3303 3227

3304 The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) that was run.3228 The ID of the [thread](/docs/api-reference/threads) that was run.

3305 3229

3306 - `type: "message_creation" or "tool_calls"`3230 - `type: "message_creation" or "tool_calls"`

3307 3231

3331 3255

3332 - `"thread.run.step.created"`3256 - `"thread.run.step.created"`

3333 3257

3334 - `ThreadRunStepInProgress object { data, event }`3258 - `object { data, event }`

3335 3259

3336 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) moves to an `in_progress` state.3260 Occurs when a [run step](/docs/api-reference/run-steps/step-object) moves to an `in_progress` state.

3337 3261

3338 - `data: RunStep`3262 - `data: RunStep`

3339 3263

3343 3267

3344 - `"thread.run.step.in_progress"`3268 - `"thread.run.step.in_progress"`

3345 3269

3346 - `ThreadRunStepDelta object { data, event }`3270 - `object { data, event }`

3347 3271

3348 Occurs when parts of a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) are being streamed.3272 Occurs when parts of a [run step](/docs/api-reference/run-steps/step-object) are being streamed.

3349 3273

3350 - `data: RunStepDeltaEvent`3274 - `data: RunStepDeltaEvent`

3351 3275

3355 3279

3356 The identifier of the run step, which can be referenced in API endpoints.3280 The identifier of the run step, which can be referenced in API endpoints.

3357 3281

3358 - `delta: RunStepDelta`3282 - `delta: object { step_details }`

3359 3283

3360 The delta containing the fields that have changed on the run step.3284 The delta containing the fields that have changed on the run step.

3361 3285

3389 3313

3390 - `"tool_calls"`3314 - `"tool_calls"`

3391 3315

3392 - `tool_calls: optional array of ToolCallDelta`3316 - `tool_calls: optional array of CodeInterpreterToolCallDelta or FileSearchToolCallDelta or FunctionToolCallDelta`

3393 3317

3394 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`.3318 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`.

3395 3319

3457 3381

3458 - `file_id: optional string`3382 - `file_id: optional string`

3459 3383

3460 The [file](https://platform.openai.com/docs/api-reference/files) ID of the image.3384 The [file](/docs/api-reference/files) ID of the image.

3461 3385

3462 - `FileSearchToolCallDelta object { file_search, index, type, id }`3386 - `FileSearchToolCallDelta object { file_search, index, type, id }`

3463 3387

3509 3433

3510 - `output: optional string`3434 - `output: optional string`

3511 3435

3512 The output of the function. This will be `null` if the outputs have not been [submitted](https://platform.openai.com/docs/api-reference/runs/submitToolOutputs) yet.3436 The output of the function. This will be `null` if the outputs have not been [submitted](/docs/api-reference/runs/submitToolOutputs) yet.

3513 3437

3514 - `object: "thread.run.step.delta"`3438 - `object: "thread.run.step.delta"`

3515 3439

3521 3445

3522 - `"thread.run.step.delta"`3446 - `"thread.run.step.delta"`

3523 3447

3524 - `ThreadRunStepCompleted object { data, event }`3448 - `object { data, event }`

3525 3449

3526 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is completed.3450 Occurs when a [run step](/docs/api-reference/run-steps/step-object) is completed.

3527 3451

3528 - `data: RunStep`3452 - `data: RunStep`

3529 3453

3533 3457

3534 - `"thread.run.step.completed"`3458 - `"thread.run.step.completed"`

3535 3459

3536 - `ThreadRunStepFailed object { data, event }`3460 - `object { data, event }`

3537 3461

3538 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) fails.3462 Occurs when a [run step](/docs/api-reference/run-steps/step-object) fails.

3539 3463

3540 - `data: RunStep`3464 - `data: RunStep`

3541 3465

3545 3469

3546 - `"thread.run.step.failed"`3470 - `"thread.run.step.failed"`

3547 3471

3548 - `ThreadRunStepCancelled object { data, event }`3472 - `object { data, event }`

3549 3473

3550 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is cancelled.3474 Occurs when a [run step](/docs/api-reference/run-steps/step-object) is cancelled.

3551 3475

3552 - `data: RunStep`3476 - `data: RunStep`

3553 3477

3557 3481

3558 - `"thread.run.step.cancelled"`3482 - `"thread.run.step.cancelled"`

3559 3483

3560 - `ThreadRunStepExpired object { data, event }`3484 - `object { data, event }`

3561 3485

3562 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) expires.3486 Occurs when a [run step](/docs/api-reference/run-steps/step-object) expires.

3563 3487

3564 - `data: RunStep`3488 - `data: RunStep`

3565 3489

3569 3493

3570 - `"thread.run.step.expired"`3494 - `"thread.run.step.expired"`

3571 3495

3572 - `ThreadMessageCreated object { data, event }`3496 - `object { data, event }`

3573 3497

3574 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.3498 Occurs when a [message](/docs/api-reference/messages/object) is created.

3575 3499

3576 - `data: Message`3500 - `data: Message`

3577 3501

3578 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).3502 Represents a message within a [thread](/docs/api-reference/threads).

3579 3503

3580 - `id: string`3504 - `id: string`

3581 3505

3583 3507

3584 - `assistant_id: string`3508 - `assistant_id: string`

3585 3509

3586 If applicable, the ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) that authored this message.3510 If applicable, the ID of the [assistant](/docs/api-reference/assistants) that authored this message.

3587 3511

3588 - `attachments: array of object { file_id, tools }`3512 - `attachments: array of object { file_id, tools }`

3589 3513

3599 3523

3600 - `CodeInterpreterTool object { type }`3524 - `CodeInterpreterTool object { type }`

3601 3525

3602 - `AssistantToolsFileSearchTypeOnly object { type }`3526 - `FileSearchTool object { type }`

3603 3527

3604 - `type: "file_search"`3528 - `type: "file_search"`

3605 3529

3611 3535

3612 The Unix timestamp (in seconds) for when the message was completed.3536 The Unix timestamp (in seconds) for when the message was completed.

3613 3537

3614 - `content: array of MessageContent`3538 - `content: array of ImageFileContentBlock or ImageURLContentBlock or TextContentBlock or RefusalContentBlock`

3615 3539

3616 The content of the message in array of text and/or images.3540 The content of the message in array of text and/or images.

3617 3541

3618 - `ImageFileContentBlock object { image_file, type }`3542 - `ImageFileContentBlock object { image_file, type }`

3619 3543

3620 References an image [File](https://platform.openai.com/docs/api-reference/files) in the content of a message.3544 References an image [File](/docs/api-reference/files) in the content of a message.

3621 3545

3622 - `image_file: ImageFile`3546 - `image_file: ImageFile`

3623 3547

3624 - `file_id: string`3548 - `file_id: string`

3625 3549

3626 The [File](https://platform.openai.com/docs/api-reference/files) ID of the image in the message content. Set `purpose="vision"` when uploading the File if you need to later display the file content.3550 The [File](/docs/api-reference/files) ID of the image in the message content. Set `purpose="vision"` when uploading the File if you need to later display the file content.

3627 3551

3628 - `detail: optional "auto" or "low" or "high"`3552 - `detail: optional "auto" or "low" or "high"`

3629 3553

3673 3597

3674 - `text: Text`3598 - `text: Text`

3675 3599

3676 - `annotations: array of Annotation`3600 - `annotations: array of FileCitationAnnotation or FilePathAnnotation`

3677 3601

3678 - `FileCitationAnnotation object { end_index, file_citation, start_index, 2 more }`3602 - `FileCitationAnnotation object { end_index, file_citation, start_index, 2 more }`

3679 3603

3796 3720

3797 - `run_id: string`3721 - `run_id: string`

3798 3722

3799 The ID of the [run](https://platform.openai.com/docs/api-reference/runs) associated with the creation of this message. Value is `null` when messages are created manually using the create message or create thread endpoints.3723 The ID of the [run](/docs/api-reference/runs) associated with the creation of this message. Value is `null` when messages are created manually using the create message or create thread endpoints.

3800 3724

3801 - `status: "in_progress" or "incomplete" or "completed"`3725 - `status: "in_progress" or "incomplete" or "completed"`

3802 3726

3810 3734

3811 - `thread_id: string`3735 - `thread_id: string`

3812 3736

3813 The [thread](https://platform.openai.com/docs/api-reference/threads) ID that this message belongs to.3737 The [thread](/docs/api-reference/threads) ID that this message belongs to.

3814 3738

3815 - `event: "thread.message.created"`3739 - `event: "thread.message.created"`

3816 3740

3817 - `"thread.message.created"`3741 - `"thread.message.created"`

3818 3742

3819 - `ThreadMessageInProgress object { data, event }`3743 - `object { data, event }`

3820 3744

3821 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) moves to an `in_progress` state.3745 Occurs when a [message](/docs/api-reference/messages/object) moves to an `in_progress` state.

3822 3746

3823 - `data: Message`3747 - `data: Message`

3824 3748

3825 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).3749 Represents a message within a [thread](/docs/api-reference/threads).

3826 3750

3827 - `event: "thread.message.in_progress"`3751 - `event: "thread.message.in_progress"`

3828 3752

3829 - `"thread.message.in_progress"`3753 - `"thread.message.in_progress"`

3830 3754

3831 - `ThreadMessageDelta object { data, event }`3755 - `object { data, event }`

3832 3756

3833 Occurs when parts of a [Message](https://platform.openai.com/docs/api-reference/messages/object) are being streamed.3757 Occurs when parts of a [Message](/docs/api-reference/messages/object) are being streamed.

3834 3758

3835 - `data: MessageDeltaEvent`3759 - `data: MessageDeltaEvent`

3836 3760

3844 3768

3845 The delta containing the fields that have changed on the Message.3769 The delta containing the fields that have changed on the Message.

3846 3770

3847 - `content: optional array of MessageContentDelta`3771 - `content: optional array of ImageFileDeltaBlock or TextDeltaBlock or RefusalDeltaBlock or ImageURLDeltaBlock`

3848 3772

3849 The content of the message in array of text and/or images.3773 The content of the message in array of text and/or images.

3850 3774

3851 - `ImageFileDeltaBlock object { index, type, image_file }`3775 - `ImageFileDeltaBlock object { index, type, image_file }`

3852 3776

3853 References an image [File](https://platform.openai.com/docs/api-reference/files) in the content of a message.3777 References an image [File](/docs/api-reference/files) in the content of a message.

3854 3778

3855 - `index: number`3779 - `index: number`

3856 3780

3876 3800

3877 - `file_id: optional string`3801 - `file_id: optional string`

3878 3802

3879 The [File](https://platform.openai.com/docs/api-reference/files) ID of the image in the message content. Set `purpose="vision"` when uploading the File if you need to later display the file content.3803 The [File](/docs/api-reference/files) ID of the image in the message content. Set `purpose="vision"` when uploading the File if you need to later display the file content.

3880 3804

3881 - `TextDeltaBlock object { index, type, text }`3805 - `TextDeltaBlock object { index, type, text }`

3882 3806

3894 3818

3895 - `text: optional TextDelta`3819 - `text: optional TextDelta`

3896 3820

3897 - `annotations: optional array of AnnotationDelta`3821 - `annotations: optional array of FileCitationDeltaAnnotation or FilePathDeltaAnnotation`

3898 3822

3899 - `FileCitationDeltaAnnotation object { index, type, end_index, 3 more }`3823 - `FileCitationDeltaAnnotation object { index, type, end_index, 3 more }`

3900 3824

4024 3948

4025 - `"thread.message.delta"`3949 - `"thread.message.delta"`

4026 3950

4027 - `ThreadMessageCompleted object { data, event }`3951 - `object { data, event }`

4028 3952

4029 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is completed.3953 Occurs when a [message](/docs/api-reference/messages/object) is completed.

4030 3954

4031 - `data: Message`3955 - `data: Message`

4032 3956

4033 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).3957 Represents a message within a [thread](/docs/api-reference/threads).

4034 3958

4035 - `event: "thread.message.completed"`3959 - `event: "thread.message.completed"`

4036 3960

4037 - `"thread.message.completed"`3961 - `"thread.message.completed"`

4038 3962

4039 - `ThreadMessageIncomplete object { data, event }`3963 - `object { data, event }`

4040 3964

4041 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) ends before it is completed.3965 Occurs when a [message](/docs/api-reference/messages/object) ends before it is completed.

4042 3966

4043 - `data: Message`3967 - `data: Message`

4044 3968

4045 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).3969 Represents a message within a [thread](/docs/api-reference/threads).

4046 3970

4047 - `event: "thread.message.incomplete"`3971 - `event: "thread.message.incomplete"`

4048 3972

4050 3974

4051 - `ErrorEvent object { data, event }`3975 - `ErrorEvent object { data, event }`

4052 3976

4053 Occurs when an [error](https://platform.openai.com/docs/guides/error-codes#api-errors) occurs. This can happen due to an internal server error or a timeout.3977 Occurs when an [error](/docs/guides/error-codes#api-errors) occurs. This can happen due to an internal server error or a timeout.

4054 3978

4055 - `data: ErrorObject`3979 - `data: ErrorObject`

4056 3980

4066 3990

4067 - `"error"`3991 - `"error"`

4068 3992

4069### Assistant Tool3993 - `DoneEvent object { data, event }`

4070 3994

4071- `AssistantTool = CodeInterpreterTool or FileSearchTool or FunctionTool`3995 Occurs when a stream ends.

4072 3996

4073 - `CodeInterpreterTool object { type }`3997 - `data: "[DONE]"`

4074 3998

4075 - `type: "code_interpreter"`3999 - `"[DONE]"`

4076 4000

4077 The type of tool being defined: `code_interpreter`4001 - `event: "done"`

4078 4002

4079 - `"code_interpreter"`4003 - `"done"`

~~4080~~

4081 - `FileSearchTool object { type, file_search }`

~~4082~~

4083 - `type: "file_search"`

~~4084~~

4085 The type of tool being defined: `file_search`

~~4086~~

4087 - `"file_search"`

~~4088~~

4089 - `file_search: optional object { max_num_results, ranking_options }`

~~4090~~

4091 Overrides for the file search tool.

~~4092~~

4093 - `max_num_results: optional number`

~~4094~~

4095 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.

~~4096~~

4097 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

~~4098~~

4099 - `ranking_options: optional object { score_threshold, ranker }`

~~4100~~

4101 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.

~~4102~~

4103 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

~~4104~~

4105 - `score_threshold: number`

~~4106~~

4107 The score threshold for the file search. All values must be a floating point number between 0 and 1.

~~4108~~

4109 - `ranker: optional "auto" or "default_2024_08_21"`

~~4110~~

4111 The ranker to use for the file search. If not specified will use the `auto` ranker.

~~4112~~

4113 - `"auto"`

~~4114~~

4115 - `"default_2024_08_21"`

~~4116~~

4117 - `FunctionTool object { function, type }`

~~4118~~

4119 - `function: FunctionDefinition`

~~4120~~

4121 - `name: string`

~~4122~~

4123 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.

~~4124~~

4125 - `description: optional string`

~~4126~~

4127 A description of what the function does, used by the model to choose when and how to call the function.

~~4128~~

4129 - `parameters: optional FunctionParameters`

~~4130~~

4131 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

~~4132~~

4133 Omitting `parameters` defines a function with an empty parameter list.

~~4134~~

4135 - `strict: optional boolean`

~~4136~~

4137 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](https://platform.openai.com/docs/guides/function-calling).

~~4138~~

4139 - `type: "function"`

~~4140~~

4141 The type of tool being defined: `function`

~~4142~~

4143 - `"function"`

4144 4004

4145### Code Interpreter Tool4005### Code Interpreter Tool

4146 4006

4170 4030

4171 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.4031 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.

4172 4032

4173 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.4033 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

4174 4034

4175 - `ranking_options: optional object { score_threshold, ranker }`4035 - `ranking_options: optional object { score_threshold, ranker }`

4176 4036

4177 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.4037 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.

4178 4038

4179 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.4039 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

4180 4040

4181 - `score_threshold: number`4041 - `score_threshold: number`

4182 4042

4206 4066

4207 - `parameters: optional FunctionParameters`4067 - `parameters: optional FunctionParameters`

4208 4068

4209 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.4069 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

4210 4070

4211 Omitting `parameters` defines a function with an empty parameter list.4071 Omitting `parameters` defines a function with an empty parameter list.

4212 4072

4213 - `strict: optional boolean`4073 - `strict: optional boolean`

4214 4074

4215 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](https://platform.openai.com/docs/guides/function-calling).4075 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](/docs/guides/function-calling).

4216 4076

4217 - `type: "function"`4077 - `type: "function"`

4218 4078

4224 4084

4225- `MessageStreamEvent = object { data, event } or object { data, event } or object { data, event } or 2 more`4085- `MessageStreamEvent = object { data, event } or object { data, event } or object { data, event } or 2 more`

4226 4086

4227 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.4087 Occurs when a [message](/docs/api-reference/messages/object) is created.

4228 4088

4229 - `ThreadMessageCreated object { data, event }`4089 - `object { data, event }`

4230 4090

4231 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.4091 Occurs when a [message](/docs/api-reference/messages/object) is created.

4232 4092

4233 - `data: Message`4093 - `data: Message`

4234 4094

4235 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4095 Represents a message within a [thread](/docs/api-reference/threads).

4236 4096

4237 - `id: string`4097 - `id: string`

4238 4098

4240 4100

4241 - `assistant_id: string`4101 - `assistant_id: string`

4242 4102

4243 If applicable, the ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) that authored this message.4103 If applicable, the ID of the [assistant](/docs/api-reference/assistants) that authored this message.

4244 4104

4245 - `attachments: array of object { file_id, tools }`4105 - `attachments: array of object { file_id, tools }`

4246 4106

4262 4122

4263 - `"code_interpreter"`4123 - `"code_interpreter"`

4264 4124

4265 - `AssistantToolsFileSearchTypeOnly object { type }`4125 - `FileSearchTool object { type }`

4266 4126

4267 - `type: "file_search"`4127 - `type: "file_search"`

4268 4128

4274 4134

4275 The Unix timestamp (in seconds) for when the message was completed.4135 The Unix timestamp (in seconds) for when the message was completed.

4276 4136

4277 - `content: array of MessageContent`4137 - `content: array of ImageFileContentBlock or ImageURLContentBlock or TextContentBlock or RefusalContentBlock`

4278 4138

4279 The content of the message in array of text and/or images.4139 The content of the message in array of text and/or images.

4280 4140

4281 - `ImageFileContentBlock object { image_file, type }`4141 - `ImageFileContentBlock object { image_file, type }`

4282 4142

4283 References an image [File](https://platform.openai.com/docs/api-reference/files) in the content of a message.4143 References an image [File](/docs/api-reference/files) in the content of a message.

4284 4144

4285 - `image_file: ImageFile`4145 - `image_file: ImageFile`

4286 4146

4287 - `file_id: string`4147 - `file_id: string`

4288 4148

4289 The [File](https://platform.openai.com/docs/api-reference/files) ID of the image in the message content. Set `purpose="vision"` when uploading the File if you need to later display the file content.4149 The [File](/docs/api-reference/files) ID of the image in the message content. Set `purpose="vision"` when uploading the File if you need to later display the file content.

4290 4150

4291 - `detail: optional "auto" or "low" or "high"`4151 - `detail: optional "auto" or "low" or "high"`

4292 4152

4336 4196

4337 - `text: Text`4197 - `text: Text`

4338 4198

4339 - `annotations: array of Annotation`4199 - `annotations: array of FileCitationAnnotation or FilePathAnnotation`

4340 4200

4341 - `FileCitationAnnotation object { end_index, file_citation, start_index, 2 more }`4201 - `FileCitationAnnotation object { end_index, file_citation, start_index, 2 more }`

4342 4202

4459 4319

4460 - `run_id: string`4320 - `run_id: string`

4461 4321

4462 The ID of the [run](https://platform.openai.com/docs/api-reference/runs) associated with the creation of this message. Value is `null` when messages are created manually using the create message or create thread endpoints.4322 The ID of the [run](/docs/api-reference/runs) associated with the creation of this message. Value is `null` when messages are created manually using the create message or create thread endpoints.

4463 4323

4464 - `status: "in_progress" or "incomplete" or "completed"`4324 - `status: "in_progress" or "incomplete" or "completed"`

4465 4325

4473 4333

4474 - `thread_id: string`4334 - `thread_id: string`

4475 4335

4476 The [thread](https://platform.openai.com/docs/api-reference/threads) ID that this message belongs to.4336 The [thread](/docs/api-reference/threads) ID that this message belongs to.

4477 4337

4478 - `event: "thread.message.created"`4338 - `event: "thread.message.created"`

4479 4339

4480 - `"thread.message.created"`4340 - `"thread.message.created"`

4481 4341

4482 - `ThreadMessageInProgress object { data, event }`4342 - `object { data, event }`

4483 4343

4484 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) moves to an `in_progress` state.4344 Occurs when a [message](/docs/api-reference/messages/object) moves to an `in_progress` state.

4485 4345

4486 - `data: Message`4346 - `data: Message`

4487 4347

4488 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4348 Represents a message within a [thread](/docs/api-reference/threads).

4489 4349

4490 - `event: "thread.message.in_progress"`4350 - `event: "thread.message.in_progress"`

4491 4351

4492 - `"thread.message.in_progress"`4352 - `"thread.message.in_progress"`

4493 4353

4494 - `ThreadMessageDelta object { data, event }`4354 - `object { data, event }`

4495 4355

4496 Occurs when parts of a [Message](https://platform.openai.com/docs/api-reference/messages/object) are being streamed.4356 Occurs when parts of a [Message](/docs/api-reference/messages/object) are being streamed.

4497 4357

4498 - `data: MessageDeltaEvent`4358 - `data: MessageDeltaEvent`

4499 4359

4507 4367

4508 The delta containing the fields that have changed on the Message.4368 The delta containing the fields that have changed on the Message.

4509 4369

4510 - `content: optional array of MessageContentDelta`4370 - `content: optional array of ImageFileDeltaBlock or TextDeltaBlock or RefusalDeltaBlock or ImageURLDeltaBlock`

4511 4371

4512 The content of the message in array of text and/or images.4372 The content of the message in array of text and/or images.

4513 4373

4514 - `ImageFileDeltaBlock object { index, type, image_file }`4374 - `ImageFileDeltaBlock object { index, type, image_file }`

4515 4375

4516 References an image [File](https://platform.openai.com/docs/api-reference/files) in the content of a message.4376 References an image [File](/docs/api-reference/files) in the content of a message.

4517 4377

4518 - `index: number`4378 - `index: number`

4519 4379

4539 4399

4540 - `file_id: optional string`4400 - `file_id: optional string`

4541 4401

4542 The [File](https://platform.openai.com/docs/api-reference/files) ID of the image in the message content. Set `purpose="vision"` when uploading the File if you need to later display the file content.4402 The [File](/docs/api-reference/files) ID of the image in the message content. Set `purpose="vision"` when uploading the File if you need to later display the file content.

4543 4403

4544 - `TextDeltaBlock object { index, type, text }`4404 - `TextDeltaBlock object { index, type, text }`

4545 4405

4557 4417

4558 - `text: optional TextDelta`4418 - `text: optional TextDelta`

4559 4419

4560 - `annotations: optional array of AnnotationDelta`4420 - `annotations: optional array of FileCitationDeltaAnnotation or FilePathDeltaAnnotation`

4561 4421

4562 - `FileCitationDeltaAnnotation object { index, type, end_index, 3 more }`4422 - `FileCitationDeltaAnnotation object { index, type, end_index, 3 more }`

4563 4423

4687 4547

4688 - `"thread.message.delta"`4548 - `"thread.message.delta"`

4689 4549

4690 - `ThreadMessageCompleted object { data, event }`4550 - `object { data, event }`

4691 4551

4692 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is completed.4552 Occurs when a [message](/docs/api-reference/messages/object) is completed.

4693 4553

4694 - `data: Message`4554 - `data: Message`

4695 4555

4696 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4556 Represents a message within a [thread](/docs/api-reference/threads).

4697 4557

4698 - `event: "thread.message.completed"`4558 - `event: "thread.message.completed"`

4699 4559

4700 - `"thread.message.completed"`4560 - `"thread.message.completed"`

4701 4561

4702 - `ThreadMessageIncomplete object { data, event }`4562 - `object { data, event }`

4703 4563

4704 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) ends before it is completed.4564 Occurs when a [message](/docs/api-reference/messages/object) ends before it is completed.

4705 4565

4706 - `data: Message`4566 - `data: Message`

4707 4567

4708 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4568 Represents a message within a [thread](/docs/api-reference/threads).

4709 4569

4710 - `event: "thread.message.incomplete"`4570 - `event: "thread.message.incomplete"`

4711 4571

4715 4575

4716- `RunStepStreamEvent = object { data, event } or object { data, event } or object { data, event } or 4 more`4576- `RunStepStreamEvent = object { data, event } or object { data, event } or object { data, event } or 4 more`

4717 4577

4718 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is created.4578 Occurs when a [run step](/docs/api-reference/run-steps/step-object) is created.

4719 4579

4720 - `ThreadRunStepCreated object { data, event }`4580 - `object { data, event }`

4721 4581

4722 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is created.4582 Occurs when a [run step](/docs/api-reference/run-steps/step-object) is created.

4723 4583

4724 - `data: RunStep`4584 - `data: RunStep`

4725 4585

4731 4591

4732 - `assistant_id: string`4592 - `assistant_id: string`

4733 4593

4734 The ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) associated with the run step.4594 The ID of the [assistant](/docs/api-reference/assistants) associated with the run step.

4735 4595

4736 - `cancelled_at: number`4596 - `cancelled_at: number`

4737 4597

4786 4646

4787 - `run_id: string`4647 - `run_id: string`

4788 4648

4789 The ID of the [run](https://platform.openai.com/docs/api-reference/runs) that this run step is a part of.4649 The ID of the [run](/docs/api-reference/runs) that this run step is a part of.

4790 4650

4791 - `status: "in_progress" or "cancelled" or "failed" or 2 more`4651 - `status: "in_progress" or "cancelled" or "failed" or 2 more`

4792 4652

4826 4686

4827 Details of the tool call.4687 Details of the tool call.

4828 4688

4829 - `tool_calls: array of ToolCall`4689 - `tool_calls: array of CodeInterpreterToolCall or FileSearchToolCall or FunctionToolCall`

4830 4690

4831 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`.4691 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`.

4832 4692

4850 4710

4851 The outputs from the Code Interpreter tool call. Code Interpreter can output one or more items, including text (`logs`) or images (`image`). Each of these are represented by a different object type.4711 The outputs from the Code Interpreter tool call. Code Interpreter can output one or more items, including text (`logs`) or images (`image`). Each of these are represented by a different object type.

4852 4712

4853 - `Logs object { logs, type }`4713 - `CodeInterpreterLogOutput object { logs, type }`

4854 4714

4855 Text output from the Code Interpreter tool call as part of a run step.4715 Text output from the Code Interpreter tool call as part of a run step.

4856 4716

4864 4724

4865 - `"logs"`4725 - `"logs"`

4866 4726

4867 - `Image object { image, type }`4727 - `CodeInterpreterImageOutput object { image, type }`

4868 4728

4869 - `image: object { file_id }`4729 - `image: object { file_id }`

4870 4730

4871 - `file_id: string`4731 - `file_id: string`

4872 4732

4873 The [file](https://platform.openai.com/docs/api-reference/files) ID of the image.4733 The [file](/docs/api-reference/files) ID of the image.

4874 4734

4875 - `type: "image"`4735 - `type: "image"`

4876 4736

4966 4826

4967 - `output: string`4827 - `output: string`

4968 4828

4969 The output of the function. This will be `null` if the outputs have not been [submitted](https://platform.openai.com/docs/api-reference/runs/submitToolOutputs) yet.4829 The output of the function. This will be `null` if the outputs have not been [submitted](/docs/api-reference/runs/submitToolOutputs) yet.

4970 4830

4971 - `type: "function"`4831 - `type: "function"`

4972 4832

4982 4842

4983 - `thread_id: string`4843 - `thread_id: string`

4984 4844

4985 The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) that was run.4845 The ID of the [thread](/docs/api-reference/threads) that was run.

4986 4846

4987 - `type: "message_creation" or "tool_calls"`4847 - `type: "message_creation" or "tool_calls"`

4988 4848

5012 4872

5013 - `"thread.run.step.created"`4873 - `"thread.run.step.created"`

5014 4874

5015 - `ThreadRunStepInProgress object { data, event }`4875 - `object { data, event }`

5016 4876

5017 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) moves to an `in_progress` state.4877 Occurs when a [run step](/docs/api-reference/run-steps/step-object) moves to an `in_progress` state.

5018 4878

5019 - `data: RunStep`4879 - `data: RunStep`

5020 4880

5024 4884

5025 - `"thread.run.step.in_progress"`4885 - `"thread.run.step.in_progress"`

5026 4886

5027 - `ThreadRunStepDelta object { data, event }`4887 - `object { data, event }`

5028 4888

5029 Occurs when parts of a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) are being streamed.4889 Occurs when parts of a [run step](/docs/api-reference/run-steps/step-object) are being streamed.

5030 4890

5031 - `data: RunStepDeltaEvent`4891 - `data: RunStepDeltaEvent`

5032 4892

5036 4896

5037 The identifier of the run step, which can be referenced in API endpoints.4897 The identifier of the run step, which can be referenced in API endpoints.

5038 4898

5039 - `delta: RunStepDelta`4899 - `delta: object { step_details }`

5040 4900

5041 The delta containing the fields that have changed on the run step.4901 The delta containing the fields that have changed on the run step.

5042 4902

5070 4930

5071 - `"tool_calls"`4931 - `"tool_calls"`

5072 4932

5073 - `tool_calls: optional array of ToolCallDelta`4933 - `tool_calls: optional array of CodeInterpreterToolCallDelta or FileSearchToolCallDelta or FunctionToolCallDelta`

5074 4934

5075 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`.4935 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`.

5076 4936

5138 4998

5139 - `file_id: optional string`4999 - `file_id: optional string`

5140 5000

5141 The [file](https://platform.openai.com/docs/api-reference/files) ID of the image.5001 The [file](/docs/api-reference/files) ID of the image.

5142 5002

5143 - `FileSearchToolCallDelta object { file_search, index, type, id }`5003 - `FileSearchToolCallDelta object { file_search, index, type, id }`

5144 5004

5190 5050

5191 - `output: optional string`5051 - `output: optional string`

5192 5052

5193 The output of the function. This will be `null` if the outputs have not been [submitted](https://platform.openai.com/docs/api-reference/runs/submitToolOutputs) yet.5053 The output of the function. This will be `null` if the outputs have not been [submitted](/docs/api-reference/runs/submitToolOutputs) yet.

5194 5054

5195 - `object: "thread.run.step.delta"`5055 - `object: "thread.run.step.delta"`

5196 5056

5202 5062

5203 - `"thread.run.step.delta"`5063 - `"thread.run.step.delta"`

5204 5064

5205 - `ThreadRunStepCompleted object { data, event }`5065 - `object { data, event }`

5206 5066

5207 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is completed.5067 Occurs when a [run step](/docs/api-reference/run-steps/step-object) is completed.

5208 5068

5209 - `data: RunStep`5069 - `data: RunStep`

5210 5070

5214 5074

5215 - `"thread.run.step.completed"`5075 - `"thread.run.step.completed"`

5216 5076

5217 - `ThreadRunStepFailed object { data, event }`5077 - `object { data, event }`

5218 5078

5219 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) fails.5079 Occurs when a [run step](/docs/api-reference/run-steps/step-object) fails.

5220 5080

5221 - `data: RunStep`5081 - `data: RunStep`

5222 5082

5226 5086

5227 - `"thread.run.step.failed"`5087 - `"thread.run.step.failed"`

5228 5088

5229 - `ThreadRunStepCancelled object { data, event }`5089 - `object { data, event }`

5230 5090

5231 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is cancelled.5091 Occurs when a [run step](/docs/api-reference/run-steps/step-object) is cancelled.

5232 5092

5233 - `data: RunStep`5093 - `data: RunStep`

5234 5094

5238 5098

5239 - `"thread.run.step.cancelled"`5099 - `"thread.run.step.cancelled"`

5240 5100

5241 - `ThreadRunStepExpired object { data, event }`5101 - `object { data, event }`

5242 5102

5243 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) expires.5103 Occurs when a [run step](/docs/api-reference/run-steps/step-object) expires.

5244 5104

5245 - `data: RunStep`5105 - `data: RunStep`

5246 5106

5254 5114

5255- `RunStreamEvent = object { data, event } or object { data, event } or object { data, event } or 7 more`5115- `RunStreamEvent = object { data, event } or object { data, event } or object { data, event } or 7 more`

5256 5116

5257 Occurs when a new [run](https://platform.openai.com/docs/api-reference/runs/object) is created.5117 Occurs when a new [run](/docs/api-reference/runs/object) is created.

5258 5118

5259 - `ThreadRunCreated object { data, event }`5119 - `object { data, event }`

5260 5120

5261 Occurs when a new [run](https://platform.openai.com/docs/api-reference/runs/object) is created.5121 Occurs when a new [run](/docs/api-reference/runs/object) is created.

5262 5122

5263 - `data: Run`5123 - `data: Run`

5264 5124

5265 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5125 Represents an execution run on a [thread](/docs/api-reference/threads).

5266 5126

5267 - `id: string`5127 - `id: string`

5268 5128

5270 5130

5271 - `assistant_id: string`5131 - `assistant_id: string`

5272 5132

5273 The ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for execution of this run.5133 The ID of the [assistant](/docs/api-reference/assistants) used for execution of this run.

5274 5134

5275 - `cancelled_at: number`5135 - `cancelled_at: number`

5276 5136

5306 5166

5307 - `instructions: string`5167 - `instructions: string`

5308 5168

5309 The instructions that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.5169 The instructions that the [assistant](/docs/api-reference/assistants) used for this run.

5310 5170

5311 - `last_error: object { code, message }`5171 - `last_error: object { code, message }`

5312 5172

5345 5205

5346 - `model: string`5206 - `model: string`

5347 5207

5348 The model that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.5208 The model that the [assistant](/docs/api-reference/assistants) used for this run.

5349 5209

5350 - `object: "thread.run"`5210 - `object: "thread.run"`

5351 5211

5355 5215

5356 - `parallel_tool_calls: boolean`5216 - `parallel_tool_calls: boolean`

5357 5217

5358 Whether to enable [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling) during tool use.5218 Whether to enable [parallel function calling](/docs/guides/function-calling#configuring-parallel-function-calling) during tool use.

5359 5219

5360 - `required_action: object { submit_tool_outputs, type }`5220 - `required_action: object { submit_tool_outputs, type }`

5361 5221

5371 5231

5372 - `id: string`5232 - `id: string`

5373 5233

5374 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](https://platform.openai.com/docs/api-reference/runs/submitToolOutputs) endpoint.5234 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](/docs/api-reference/runs/submitToolOutputs) endpoint.

5375 5235

5376 - `function: object { arguments, name }`5236 - `function: object { arguments, name }`

5377 5237

5399 5259

5400 - `response_format: AssistantResponseFormatOption`5260 - `response_format: AssistantResponseFormatOption`

5401 5261

5402 Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.5262 Specifies the format that the model must output. Compatible with [GPT-4o](/docs/models#gpt-4o), [GPT-4 Turbo](/docs/models#gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

5403 5263

5404 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](https://platform.openai.com/docs/guides/structured-outputs).5264 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](/docs/guides/structured-outputs).

5405 5265

5406 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.5266 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

5407 5267

5439 - `ResponseFormatJSONSchema object { json_schema, type }`5299 - `ResponseFormatJSONSchema object { json_schema, type }`

5440 5300

5441 JSON Schema response format. Used to generate structured JSON responses.5301 JSON Schema response format. Used to generate structured JSON responses.

5442 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).5302 Learn more about [Structured Outputs](/docs/guides/structured-outputs).

5443 5303

5444 - `json_schema: object { name, description, schema, strict }`5304 - `json_schema: object { name, description, schema, strict }`

5445 5305

5466 If set to true, the model will always follow the exact schema defined5326 If set to true, the model will always follow the exact schema defined

5467 in the `schema` field. Only a subset of JSON Schema is supported when5327 in the `schema` field. Only a subset of JSON Schema is supported when

5468 `strict` is `true`. To learn more, read the [Structured Outputs5328 `strict` is `true`. To learn more, read the [Structured Outputs

5469 guide](https://platform.openai.com/docs/guides/structured-outputs).5329 guide](/docs/guides/structured-outputs).

5470 5330

5471 - `type: "json_schema"`5331 - `type: "json_schema"`

5472 5332

5478 5338

5479 The Unix timestamp (in seconds) for when the run was started.5339 The Unix timestamp (in seconds) for when the run was started.

5480 5340

5481 - `status: RunStatus`5341 - `status: "queued" or "in_progress" or "requires_action" or 6 more`

5482 5342

5483 The status of the run, which can be either `queued`, `in_progress`, `requires_action`, `cancelling`, `cancelled`, `failed`, `completed`, `incomplete`, or `expired`.5343 The status of the run, which can be either `queued`, `in_progress`, `requires_action`, `cancelling`, `cancelled`, `failed`, `completed`, `incomplete`, or `expired`.

5484 5344

5502 5362

5503 - `thread_id: string`5363 - `thread_id: string`

5504 5364

5505 The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) that was executed on as a part of this run.5365 The ID of the [thread](/docs/api-reference/threads) that was executed on as a part of this run.

5506 5366

5507 - `tool_choice: AssistantToolChoiceOption`5367 - `tool_choice: AssistantToolChoiceOption`

5508 5368

5512 `required` means the model must call one or more tools before responding to the user.5372 `required` means the model must call one or more tools before responding to the user.

5513 Specifying a particular tool like `{"type": "file_search"}` or `{"type": "function", "function": {"name": "my_function"}}` forces the model to call that tool.5373 Specifying a particular tool like `{"type": "file_search"}` or `{"type": "function", "function": {"name": "my_function"}}` forces the model to call that tool.

5514 5374

5515 - `Auto = "none" or "auto" or "required"`5375 - `"none" or "auto" or "required"`

5516 5376

5517 `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.5377 `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.

5518 5378

5542 5402

5543 The name of the function to call.5403 The name of the function to call.

5544 5404

5545 - `tools: array of AssistantTool`5405 - `tools: array of CodeInterpreterTool or FileSearchTool or FunctionTool`

5546 5406

5547 The list of tools that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.5407 The list of tools that the [assistant](/docs/api-reference/assistants) used for this run.

5548 5408

5549 - `CodeInterpreterTool object { type }`5409 - `CodeInterpreterTool object { type }`

5550 5410

5570 5430

5571 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.5431 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.

5572 5432

5573 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.5433 Note that the file search tool may output fewer than `max_num_results` results. See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

5574 5434

5575 - `ranking_options: optional object { score_threshold, ranker }`5435 - `ranking_options: optional object { score_threshold, ranker }`

5576 5436

5577 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.5437 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.

5578 5438

5579 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.5439 See the [file search tool documentation](/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

5580 5440

5581 - `score_threshold: number`5441 - `score_threshold: number`

5582 5442

5604 5464

5605 - `parameters: optional FunctionParameters`5465 - `parameters: optional FunctionParameters`

5606 5466

5607 The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.5467 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

5608 5468

5609 Omitting `parameters` defines a function with an empty parameter list.5469 Omitting `parameters` defines a function with an empty parameter list.

5610 5470

5611 - `strict: optional boolean`5471 - `strict: optional boolean`

5612 5472

5613 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](https://platform.openai.com/docs/guides/function-calling).5473 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](/docs/guides/function-calling).

5614 5474

5615 - `type: "function"`5475 - `type: "function"`

5616 5476

5662 5522

5663 - `"thread.run.created"`5523 - `"thread.run.created"`

5664 5524

5665 - `ThreadRunQueued object { data, event }`5525 - `object { data, event }`

5666 5526

5667 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) moves to a `queued` status.5527 Occurs when a [run](/docs/api-reference/runs/object) moves to a `queued` status.

5668 5528

5669 - `data: Run`5529 - `data: Run`

5670 5530

5671 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5531 Represents an execution run on a [thread](/docs/api-reference/threads).

5672 5532

5673 - `event: "thread.run.queued"`5533 - `event: "thread.run.queued"`

5674 5534

5675 - `"thread.run.queued"`5535 - `"thread.run.queued"`

5676 5536

5677 - `ThreadRunInProgress object { data, event }`5537 - `object { data, event }`

5678 5538

5679 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) moves to an `in_progress` status.5539 Occurs when a [run](/docs/api-reference/runs/object) moves to an `in_progress` status.

5680 5540

5681 - `data: Run`5541 - `data: Run`

5682 5542

5683 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5543 Represents an execution run on a [thread](/docs/api-reference/threads).

5684 5544

5685 - `event: "thread.run.in_progress"`5545 - `event: "thread.run.in_progress"`

5686 5546

5687 - `"thread.run.in_progress"`5547 - `"thread.run.in_progress"`

5688 5548

5689 - `ThreadRunRequiresAction object { data, event }`5549 - `object { data, event }`

5690 5550

5691 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) moves to a `requires_action` status.5551 Occurs when a [run](/docs/api-reference/runs/object) moves to a `requires_action` status.

5692 5552

5693 - `data: Run`5553 - `data: Run`

5694 5554

5695 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5555 Represents an execution run on a [thread](/docs/api-reference/threads).

5696 5556

5697 - `event: "thread.run.requires_action"`5557 - `event: "thread.run.requires_action"`

5698 5558

5699 - `"thread.run.requires_action"`5559 - `"thread.run.requires_action"`

5700 5560

5701 - `ThreadRunCompleted object { data, event }`5561 - `object { data, event }`

5702 5562

5703 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) is completed.5563 Occurs when a [run](/docs/api-reference/runs/object) is completed.

5704 5564

5705 - `data: Run`5565 - `data: Run`

5706 5566

5707 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5567 Represents an execution run on a [thread](/docs/api-reference/threads).

5708 5568

5709 - `event: "thread.run.completed"`5569 - `event: "thread.run.completed"`

5710 5570

5711 - `"thread.run.completed"`5571 - `"thread.run.completed"`

5712 5572

5713 - `ThreadRunIncomplete object { data, event }`5573 - `object { data, event }`

5714 5574

5715 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) ends with status `incomplete`.5575 Occurs when a [run](/docs/api-reference/runs/object) ends with status `incomplete`.

5716 5576

5717 - `data: Run`5577 - `data: Run`

5718 5578

5719 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5579 Represents an execution run on a [thread](/docs/api-reference/threads).

5720 5580

5721 - `event: "thread.run.incomplete"`5581 - `event: "thread.run.incomplete"`

5722 5582

5723 - `"thread.run.incomplete"`5583 - `"thread.run.incomplete"`

5724 5584

5725 - `ThreadRunFailed object { data, event }`5585 - `object { data, event }`

5726 5586

5727 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) fails.5587 Occurs when a [run](/docs/api-reference/runs/object) fails.

5728 5588

5729 - `data: Run`5589 - `data: Run`

5730 5590

5731 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5591 Represents an execution run on a [thread](/docs/api-reference/threads).

5732 5592

5733 - `event: "thread.run.failed"`5593 - `event: "thread.run.failed"`

5734 5594

5735 - `"thread.run.failed"`5595 - `"thread.run.failed"`

5736 5596

5737 - `ThreadRunCancelling object { data, event }`5597 - `object { data, event }`

5738 5598

5739 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) moves to a `cancelling` status.5599 Occurs when a [run](/docs/api-reference/runs/object) moves to a `cancelling` status.

5740 5600

5741 - `data: Run`5601 - `data: Run`

5742 5602

5743 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5603 Represents an execution run on a [thread](/docs/api-reference/threads).

5744 5604

5745 - `event: "thread.run.cancelling"`5605 - `event: "thread.run.cancelling"`

5746 5606

5747 - `"thread.run.cancelling"`5607 - `"thread.run.cancelling"`

5748 5608

5749 - `ThreadRunCancelled object { data, event }`5609 - `object { data, event }`

5750 5610

5751 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) is cancelled.5611 Occurs when a [run](/docs/api-reference/runs/object) is cancelled.

5752 5612

5753 - `data: Run`5613 - `data: Run`

5754 5614

5755 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5615 Represents an execution run on a [thread](/docs/api-reference/threads).

5756 5616

5757 - `event: "thread.run.cancelled"`5617 - `event: "thread.run.cancelled"`

5758 5618

5759 - `"thread.run.cancelled"`5619 - `"thread.run.cancelled"`

5760 5620

5761 - `ThreadRunExpired object { data, event }`5621 - `object { data, event }`

5762 5622

5763 Occurs when a [run](https://platform.openai.com/docs/api-reference/runs/object) expires.5623 Occurs when a [run](/docs/api-reference/runs/object) expires.

5764 5624

5765 - `data: Run`5625 - `data: Run`

5766 5626

5767 Represents an execution run on a [thread](https://platform.openai.com/docs/api-reference/threads).5627 Represents an execution run on a [thread](/docs/api-reference/threads).

5768 5628

5769 - `event: "thread.run.expired"`5629 - `event: "thread.run.expired"`

5770 5630

5774 5634

5775- `ThreadStreamEvent object { data, event, enabled }`5635- `ThreadStreamEvent object { data, event, enabled }`

5776 5636

5777 Occurs when a new [thread](https://platform.openai.com/docs/api-reference/threads/object) is created.5637 Occurs when a new [thread](/docs/api-reference/threads/object) is created.

5778 5638

5779 - `data: Thread`5639 - `data: Thread`

5780 5640

5781 Represents a thread that contains [messages](https://platform.openai.com/docs/api-reference/messages).5641 Represents a thread that contains [messages](/docs/api-reference/messages).

5782 5642

5783 - `id: string`5643 - `id: string`

5784 5644

5811 5671

5812 - `file_ids: optional array of string`5672 - `file_ids: optional array of string`

5813 5673

5814 A list of [file](https://platform.openai.com/docs/api-reference/files) IDs made available to the `code_interpreter` tool. There can be a maximum of 20 files associated with the tool.5674 A list of [file](/docs/api-reference/files) IDs made available to the `code_interpreter` tool. There can be a maximum of 20 files associated with the tool.

5815 5675

5816 - `file_search: optional object { vector_store_ids }`5676 - `file_search: optional object { vector_store_ids }`

5817 5677

5818 - `vector_store_ids: optional array of string`5678 - `vector_store_ids: optional array of string`

5819 5679

5820 The [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object) attached to this thread. There can be a maximum of 1 vector store attached to the thread.5680 The [vector store](/docs/api-reference/vector-stores/object) attached to this thread. There can be a maximum of 1 vector store attached to the thread.

5821 5681

5822 - `event: "thread.created"`5682 - `event: "thread.created"`

5823 5683

resources/beta/subresources/assistants/index.md 2026-07-10 23:02 UTC to 2026-07-12 06:58 UTC

Assistants

List assistants

Query Parameters

Returns

Example

Response

Example

Response

Create assistant

Body Parameters

Returns

Example

Response

Code Interpreter

Response

Files

Response

Retrieve assistant

Path Parameters

Returns

Example

Response

Example

Response

Modify assistant

Path Parameters

Body Parameters

Returns

Example

Response

Example

Response

Delete assistant

Path Parameters

Returns

Example

Response

Example

Response

Domain Types

Assistant

Assistant Deleted

Assistant Stream Event

Code Interpreter Tool

File Search Tool

Function Tool

Message Stream Event

Run Step Stream Event

Run Stream Event

Thread Stream Event