Create assistant

POST /assistants

Create an assistant with a model and instructions.

application/json

Body Required

model string Required

Any of:
string-1 string string-2 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.

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.

Values are gpt-4o, gpt-4o-2024-05-13, 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, or gpt-3.5-turbo-16k-0613.
name string | null

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

Maximum length is 256.
description string | null

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

Maximum length is 512.
instructions string | null

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

Maximum length is 256000.
tools array[object]

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

Not more than 128 elements. Default value is [] (empty).
One of:
Code interpreter tool object FileSearch tool object Function tool object
Hide attribute Show attribute

type string Required

The type of tool being defined: code_interpreter

Value is code_interpreter.
Hide attributes Show attributes

type string Required

The type of tool being defined: file_search

Value is file_search.

file_search object

Overrides for the file search tool.

Hide file_search attribute Show file_search attribute object

max_num_results integer

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.

Minimum value is 1, maximum value is 50.
Hide attributes Show attributes

type string Required

The type of tool being defined: function

Value is function.

function object Required

Hide function attributes Show function attributes object

description string

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

name string Required

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.

parameters object

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.

Hide parameters attribute Show parameters attribute object

Additional properties are allowed
tool_resources object | null

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.
Hide tool_resources attributes Show tool_resources attributes object | null
- code_interpreter object
  Hide code_interpreter attribute Show code_interpreter attribute object
  
  file_ids array[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.
  
  Not more than 20 elements. Default value is [] (empty).
- file_search object
  
  One of:
  object-1 object object-2 object
  
  Hide attributes Show attributes
  
  vector_store_ids array[string] Required
  
  The vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.
  
  Not more than 1 element.
  
  vector_stores array[object]
  
  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.
  
  Not more than 1 element.
  
  Hide vector_stores attributes Show vector_stores attributes object
  
  file_ids array[string]
  
  A list of file IDs to add to the vector store. There can be a maximum of 10000 files in a vector store.
  
  Not more than 10000 elements.
  
  chunking_strategy object
  
  One of:
  Auto Chunking Strategy object Static Chunking Strategy object
  
  The chunking strategy used to chunk the file(s). If not set, will use the auto strategy.
  
  Hide attributes Show attributes
  
  type string Required
  
  Always static.
  
  Value is static.
  
  static object Required
  
  Hide static attributes Show static attributes object
  
  max_chunk_size_tokens integer Required
  
  The maximum number of tokens in each chunk. The default value is 800. The minimum value is 100 and the maximum value is 4096.
  
  Minimum value is 100, maximum value is 4096.
  
  chunk_overlap_tokens integer Required
  
  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.
  
  Additional properties are NOT allowed
  
  Additional properties are NOT allowed
  
  metadata object
  
  Set of 16 key-value pairs that can be attached to a vector store. This can be useful for storing additional information about the vector store in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.
  
  Hide attributes Show attributes
  
  vector_store_ids array[string]
  
  The vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.
  
  Not more than 1 element.
  
  vector_stores array[object] Required
  
  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.
  
  Not more than 1 element.
  
  Hide vector_stores attributes Show vector_stores attributes object
  
  file_ids array[string]
  
  A list of file IDs to add to the vector store. There can be a maximum of 10000 files in a vector store.
  
  Not more than 10000 elements.
  
  chunking_strategy object
  
  One of:
  Auto Chunking Strategy object Static Chunking Strategy object
  
  The chunking strategy used to chunk the file(s). If not set, will use the auto strategy.
  
  Hide attributes Show attributes
  
  type string Required
  
  Always static.
  
  Value is static.
  
  static object Required
  
  Hide static attributes Show static attributes object
  
  max_chunk_size_tokens integer Required
  
  The maximum number of tokens in each chunk. The default value is 800. The minimum value is 100 and the maximum value is 4096.
  
  Minimum value is 100, maximum value is 4096.
  
  chunk_overlap_tokens integer Required
  
  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.
  
  Additional properties are NOT allowed
  
  Additional properties are NOT allowed
  
  metadata object
  
  Set of 16 key-value pairs that can be attached to a vector store. This can be useful for storing additional information about the vector store in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.
metadata object | null

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. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.
temperature number | null

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.

Minimum value is 0, maximum value is 2. Default value is 1.
top_p number | null

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.

Minimum value is 0, maximum value is 1. Default value is 1.
response_format string | null | object
One of:
string-1 string | null AssistantsApiResponseFormat object | null

auto is the default value

Values are none or auto.
An object describing the expected output of the model. If json_object only function type tools are allowed to be passed to the Run. If text the model can return text or any value needed.
Hide attribute Show attribute

type string

Must be one of text or json_object.

Values are text or json_object. Default value is text.
Additional properties are NOT allowed

Responses

200 application/json

OK
Hide response attributes Show response attributes object
- id string Required
  
  The identifier, which can be referenced in API endpoints.
- object string Required
  
  The object type, which is always assistant.
  
  Value is assistant.
- created_at integer Required
  
  The Unix timestamp (in seconds) for when the assistant was created.
- name string | null Required
  
  The name of the assistant. The maximum length is 256 characters.
  
  Maximum length is 256.
- description string | null Required
  
  The description of the assistant. The maximum length is 512 characters.
  
  Maximum length is 512.
- model string Required
  
  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.
- instructions string | null Required
  
  The system instructions that the assistant uses. The maximum length is 256,000 characters.
  
  Maximum length is 256000.
- tools array[object] Required
  
  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.
  
  Not more than 128 elements. Default value is [] (empty).
  
  One of:
  Code interpreter tool object FileSearch tool object Function tool object
  
  Hide attributes Show attributes
  
  type string Required
  
  The type of tool being defined: file_search
  
  Value is file_search.
  
  file_search object
  
  Overrides for the file search tool.
  
  Hide file_search attribute Show file_search attribute object
  
  max_num_results integer
  
  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.
  
  Minimum value is 1, maximum value is 50.
  
  Hide attributes Show attributes
  
  type string Required
  
  The type of tool being defined: function
  
  Value is function.
  
  function object Required
  
  Hide function attributes Show function attributes object
  
  description string
  
  A description of what the function does, used by the model to choose when and how to call the function.
  
  name string Required
  
  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.
  
  parameters object
  
  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.
  
  Hide parameters attribute Show parameters attribute object
  
  Additional properties are allowed
- tool_resources object | null
  
  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.
  
  Hide tool_resources attributes Show tool_resources attributes object | null
  
  code_interpreter object
  
  Hide code_interpreter attribute Show code_interpreter attribute object
  
  file_ids array[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.
  
  Not more than 20 elements. Default value is [] (empty).
  
  file_search object
  
  Hide file_search attribute Show file_search attribute object
  
  vector_store_ids array[string]
  
  The ID of the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.
  
  Not more than 1 element.
- metadata object | null Required
  
  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. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.
- temperature number | null
  
  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.
  
  Minimum value is 0, maximum value is 2. Default value is 1.
- top_p number | null
  
  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.
  
  Minimum value is 0, maximum value is 1. Default value is 1.
- response_format string | null | object
  
  One of:
  string-1 string | null AssistantsApiResponseFormat object | null
  
  auto is the default value
  
  Values are none or auto.
  
  An object describing the expected output of the model. If json_object only function type tools are allowed to be passed to the Run. If text the model can return text or any value needed.
  
  Hide attribute Show attribute
  
  type string
  
  Must be one of text or json_object.
  
  Values are text or json_object. Default value is text.

POST /assistants

curl \
 -X POST https://api.openai.com/v1/assistants \
 -H "Authorization: Bearer $ACCESS_TOKEN" \
 -H "Content-Type: application/json" \
 -d '{"model":"gpt-4-turbo","name":"string","description":"string","instructions":"string","tools":[{"type":"code_interpreter"}],"tool_resources":{"code_interpreter":{"file_ids":[]},"file_search":{"vector_store_ids":["string"],"vector_stores":[{"file_ids":["string"],"chunking_strategy":{"type":"auto"},"metadata":{}}]}},"metadata":{},"temperature":1,"top_p":1,"response_format":"none"}'

Request example

{
  "model": "gpt-4-turbo",
  "name": "string",
  "description": "string",
  "instructions": "string",
  "tools": [
    {
      "type": "code_interpreter"
    }
  ],
  "tool_resources": {
    "code_interpreter": {
      "file_ids": []
    },
    "file_search": {
      "vector_store_ids": [
        "string"
      ],
      "vector_stores": [
        {
          "file_ids": [
            "string"
          ],
          "chunking_strategy": {
            "type": "auto"
          },
          "metadata": {}
        }
      ]
    }
  },
  "metadata": {},
  "temperature": 1,
  "top_p": 1,
  "response_format": "none"
}

Response examples (200)

{
  "id": "string",
  "object": "assistant",
  "created_at": 42,
  "name": "string",
  "description": "string",
  "model": "string",
  "instructions": "string",
  "tools": [
    {
      "type": "code_interpreter"
    }
  ],
  "tool_resources": {
    "code_interpreter": {
      "file_ids": []
    },
    "file_search": {
      "vector_store_ids": [
        "string"
      ]
    }
  },
  "metadata": {},
  "temperature": 1,
  "top_p": 1,
  "response_format": "none"
}

Create assistant

Body Required

model string Required

file_search object

response_format string | null | object

Responses

response_format string | null | object