Assistants

List assistants

beta.assistants.list(**kwargs) -> CursorPage<Assistant>

get /assistants

Returns a list of assistants.

Parameters

after: 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: 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: Integer

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

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

Returns

class Assistant

Represents an assistant that can call the model and use tools.
- id: String
  
  The identifier, which can be referenced in API endpoints.
- created_at: Integer
  
  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[AssistantTool]
  
  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.
  - class CodeInterpreterTool
    - type: :code_interpreter
      
      The type of tool being defined: code_interpreter
      - :code_interpreter
  - class FileSearchTool
    - type: :file_search
      
      The type of tool being defined: file_search
      - :file_search
    - file_search: FileSearch{ max_num_results, ranking_options}
      
      Overrides for the file search tool.
      - 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.
      - ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
  - class FunctionTool
    - 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: String
        
        A description of what the function does, used by the model to choose when and how to call the function.
      - parameters: 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: bool
        
        Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
    - type: :function
      
      The type of tool being defined: function
      - :function
- 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.
  - AssistantResponseFormatOption = :auto
    
    auto is the default value
    - :auto
  - class ResponseFormatText
    
    Default response format. Used to generate text responses.
    - type: :text
      
      The type of response format being defined. Always text.
      - :text
  - class ResponseFormatJSONObject
    
    JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
    - type: :json_object
      
      The type of response format being defined. Always json_object.
      - :json_object
  - class ResponseFormatJSONSchema
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: JSONSchema{ 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: String
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: Hash[Symbol, untyped]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: bool
        
        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
    - type: :json_schema
      
      The type of response format being defined. Always json_schema.
      - :json_schema
- temperature: Float
  
  What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
- tool_resources: ToolResources{ 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: CodeInterpreter{ file_ids}
    - 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.
  - file_search: FileSearch{ vector_store_ids}
    - 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.
- top_p: Float
  
  An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
  
  We generally recommend altering this or temperature but not both.

Example

require "openai"

openai = OpenAI::Client.new(api_key: "My API Key")

page = openai.beta.assistants.list

puts(page)

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"
}

Create assistant

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

post /assistants

Create an assistant with a model and instructions.

Parameters

model: String | ChatModel

ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.
- String = String
- ChatModel = :"gpt-5.6-sol" | :"gpt-5.6-terra" | :"gpt-5.6-luna" | 78 more
  - :"gpt-5.6-sol"
  - :"gpt-5.6-terra"
  - :"gpt-5.6-luna"
  - :"gpt-5.4"
  - :"gpt-5.4-mini"
  - :"gpt-5.4-nano"
  - :"gpt-5.4-mini-2026-03-17"
  - :"gpt-5.4-nano-2026-03-17"
  - :"gpt-5.3-chat-latest"
  - :"gpt-5.2"
  - :"gpt-5.2-2025-12-11"
  - :"gpt-5.2-chat-latest"
  - :"gpt-5.2-pro"
  - :"gpt-5.2-pro-2025-12-11"
  - :"gpt-5.1"
  - :"gpt-5.1-2025-11-13"
  - :"gpt-5.1-codex"
  - :"gpt-5.1-mini"
  - :"gpt-5.1-chat-latest"
  - :"gpt-5"
  - :"gpt-5-mini"
  - :"gpt-5-nano"
  - :"gpt-5-2025-08-07"
  - :"gpt-5-mini-2025-08-07"
  - :"gpt-5-nano-2025-08-07"
  - :"gpt-5-chat-latest"
  - :"gpt-4.1"
  - :"gpt-4.1-mini"
  - :"gpt-4.1-nano"
  - :"gpt-4.1-2025-04-14"
  - :"gpt-4.1-mini-2025-04-14"
  - :"gpt-4.1-nano-2025-04-14"
  - :"o4-mini"
  - :"o4-mini-2025-04-16"
  - :o3
  - :"o3-2025-04-16"
  - :"o3-mini"
  - :"o3-mini-2025-01-31"
  - :o1
  - :"o1-2024-12-17"
  - :"o1-preview"
  - :"o1-preview-2024-09-12"
  - :"o1-mini"
  - :"o1-mini-2024-09-12"
  - :"gpt-4o"
  - :"gpt-4o-2024-11-20"
  - :"gpt-4o-2024-08-06"
  - :"gpt-4o-2024-05-13"
  - :"gpt-4o-audio-preview"
  - :"gpt-4o-audio-preview-2024-10-01"
  - :"gpt-4o-audio-preview-2024-12-17"
  - :"gpt-4o-audio-preview-2025-06-03"
  - :"gpt-4o-mini-audio-preview"
  - :"gpt-4o-mini-audio-preview-2024-12-17"
  - :"gpt-4o-search-preview"
  - :"gpt-4o-mini-search-preview"
  - :"gpt-4o-search-preview-2025-03-11"
  - :"gpt-4o-mini-search-preview-2025-03-11"
  - :"chatgpt-4o-latest"
  - :"codex-mini-latest"
  - :"gpt-4o-mini"
  - :"gpt-4o-mini-2024-07-18"
  - :"gpt-4-turbo"
  - :"gpt-4-turbo-2024-04-09"
  - :"gpt-4-0125-preview"
  - :"gpt-4-turbo-preview"
  - :"gpt-4-1106-preview"
  - :"gpt-4-vision-preview"
  - :"gpt-4"
  - :"gpt-4-0314"
  - :"gpt-4-0613"
  - :"gpt-4-32k"
  - :"gpt-4-32k-0314"
  - :"gpt-4-32k-0613"
  - :"gpt-3.5-turbo"
  - :"gpt-3.5-turbo-16k"
  - :"gpt-3.5-turbo-0301"
  - :"gpt-3.5-turbo-0613"
  - :"gpt-3.5-turbo-1106"
  - :"gpt-3.5-turbo-0125"
  - :"gpt-3.5-turbo-16k-0613"
description: 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.
name: String

The name of the assistant. The maximum length is 256 characters.
reasoning_effort: 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: 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.
- AssistantResponseFormatOption = :auto
  
  auto is the default value
  - :auto
- class ResponseFormatText
  
  Default response format. Used to generate text responses.
  - type: :text
    
    The type of response format being defined. Always text.
    - :text
- class ResponseFormatJSONObject
  
  JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
  - type: :json_object
    
    The type of response format being defined. Always json_object.
    - :json_object
- class ResponseFormatJSONSchema
  
  JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
  - json_schema: JSONSchema{ 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: String
      
      A description of what the response format is for, used by the model to determine how to respond in the format.
    - schema: Hash[Symbol, untyped]
      
      The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
    - strict: bool
      
      Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
  - type: :json_schema
    
    The type of response format being defined. Always json_schema.
    - :json_schema
temperature: Float

What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
tool_resources: ToolResources{ 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: CodeInterpreter{ file_ids}
  - 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.
- file_search: FileSearch{ vector_store_ids, vector_stores}
  - 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.
  - vector_stores: Array[VectorStore{ 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: Auto{ type} | Static{ static, type}
      
      The chunking strategy used to chunk the file(s). If not set, will use the auto strategy.
      - class Auto
        
        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
      - class Static
        
        static: Static{ chunk_overlap_tokens, max_chunk_size_tokens}
        
        chunk_overlap_tokens: Integer
        
        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: Integer
        
        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: Array[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: 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: Array[AssistantTool]

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.
- class CodeInterpreterTool
  - type: :code_interpreter
    
    The type of tool being defined: code_interpreter
    - :code_interpreter
- class FileSearchTool
  - type: :file_search
    
    The type of tool being defined: file_search
    - :file_search
  - file_search: FileSearch{ max_num_results, ranking_options}
    
    Overrides for the file search tool.
    - 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.
    - ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
      - ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
- class FunctionTool
  - 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: String
      
      A description of what the function does, used by the model to choose when and how to call the function.
    - parameters: 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: bool
      
      Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
  - type: :function
    
    The type of tool being defined: function
    - :function
top_p: Float

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

We generally recommend altering this or temperature but not both.

Returns

class Assistant

Represents an assistant that can call the model and use tools.
- id: String
  
  The identifier, which can be referenced in API endpoints.
- created_at: Integer
  
  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[AssistantTool]
  
  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.
  - class CodeInterpreterTool
    - type: :code_interpreter
      
      The type of tool being defined: code_interpreter
      - :code_interpreter
  - class FileSearchTool
    - type: :file_search
      
      The type of tool being defined: file_search
      - :file_search
    - file_search: FileSearch{ max_num_results, ranking_options}
      
      Overrides for the file search tool.
      - 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.
      - ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
  - class FunctionTool
    - 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: String
        
        A description of what the function does, used by the model to choose when and how to call the function.
      - parameters: 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: bool
        
        Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
    - type: :function
      
      The type of tool being defined: function
      - :function
- 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.
  - AssistantResponseFormatOption = :auto
    
    auto is the default value
    - :auto
  - class ResponseFormatText
    
    Default response format. Used to generate text responses.
    - type: :text
      
      The type of response format being defined. Always text.
      - :text
  - class ResponseFormatJSONObject
    
    JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
    - type: :json_object
      
      The type of response format being defined. Always json_object.
      - :json_object
  - class ResponseFormatJSONSchema
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: JSONSchema{ 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: String
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: Hash[Symbol, untyped]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: bool
        
        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
    - type: :json_schema
      
      The type of response format being defined. Always json_schema.
      - :json_schema
- temperature: Float
  
  What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
- tool_resources: ToolResources{ 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: CodeInterpreter{ file_ids}
    - 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.
  - file_search: FileSearch{ vector_store_ids}
    - 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.
- top_p: Float
  
  An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
  
  We generally recommend altering this or temperature but not both.

Example

require "openai"

openai = OpenAI::Client.new(api_key: "My API Key")

assistant = openai.beta.assistants.create(model: :"gpt-4o")

puts(assistant)

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
}

Retrieve assistant

beta.assistants.retrieve(assistant_id) -> Assistant

get /assistants/{assistant_id}

Retrieves an assistant.

Parameters

assistant_id: String

Returns

class Assistant

Represents an assistant that can call the model and use tools.
- id: String
  
  The identifier, which can be referenced in API endpoints.
- created_at: Integer
  
  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[AssistantTool]
  
  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.
  - class CodeInterpreterTool
    - type: :code_interpreter
      
      The type of tool being defined: code_interpreter
      - :code_interpreter
  - class FileSearchTool
    - type: :file_search
      
      The type of tool being defined: file_search
      - :file_search
    - file_search: FileSearch{ max_num_results, ranking_options}
      
      Overrides for the file search tool.
      - 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.
      - ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
  - class FunctionTool
    - 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: String
        
        A description of what the function does, used by the model to choose when and how to call the function.
      - parameters: 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: bool
        
        Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
    - type: :function
      
      The type of tool being defined: function
      - :function
- 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.
  - AssistantResponseFormatOption = :auto
    
    auto is the default value
    - :auto
  - class ResponseFormatText
    
    Default response format. Used to generate text responses.
    - type: :text
      
      The type of response format being defined. Always text.
      - :text
  - class ResponseFormatJSONObject
    
    JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
    - type: :json_object
      
      The type of response format being defined. Always json_object.
      - :json_object
  - class ResponseFormatJSONSchema
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: JSONSchema{ 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: String
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: Hash[Symbol, untyped]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: bool
        
        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
    - type: :json_schema
      
      The type of response format being defined. Always json_schema.
      - :json_schema
- temperature: Float
  
  What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
- tool_resources: ToolResources{ 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: CodeInterpreter{ file_ids}
    - 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.
  - file_search: FileSearch{ vector_store_ids}
    - 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.
- top_p: Float
  
  An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
  
  We generally recommend altering this or temperature but not both.

Example

require "openai"

openai = OpenAI::Client.new(api_key: "My API Key")

assistant = openai.beta.assistants.retrieve("assistant_id")

puts(assistant)

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
}

Modify assistant

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

post /assistants/{assistant_id}

Modifies an assistant.

Parameters

assistant_id: String
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 | :"gpt-5" | :"gpt-5-mini" | :"gpt-5-nano" | 39 more

ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.
- String = String
- Model = :"gpt-5" | :"gpt-5-mini" | :"gpt-5-nano" | 39 more
  
  ID of the model to use. You can use the List models API to see all of your available models, or see our Model overview for descriptions of them.
  - :"gpt-5"
  - :"gpt-5-mini"
  - :"gpt-5-nano"
  - :"gpt-5-2025-08-07"
  - :"gpt-5-mini-2025-08-07"
  - :"gpt-5-nano-2025-08-07"
  - :"gpt-4.1"
  - :"gpt-4.1-mini"
  - :"gpt-4.1-nano"
  - :"gpt-4.1-2025-04-14"
  - :"gpt-4.1-mini-2025-04-14"
  - :"gpt-4.1-nano-2025-04-14"
  - :"o3-mini"
  - :"o3-mini-2025-01-31"
  - :o1
  - :"o1-2024-12-17"
  - :"gpt-4o"
  - :"gpt-4o-2024-11-20"
  - :"gpt-4o-2024-08-06"
  - :"gpt-4o-2024-05-13"
  - :"gpt-4o-mini"
  - :"gpt-4o-mini-2024-07-18"
  - :"gpt-4.5-preview"
  - :"gpt-4.5-preview-2025-02-27"
  - :"gpt-4-turbo"
  - :"gpt-4-turbo-2024-04-09"
  - :"gpt-4-0125-preview"
  - :"gpt-4-turbo-preview"
  - :"gpt-4-1106-preview"
  - :"gpt-4-vision-preview"
  - :"gpt-4"
  - :"gpt-4-0314"
  - :"gpt-4-0613"
  - :"gpt-4-32k"
  - :"gpt-4-32k-0314"
  - :"gpt-4-32k-0613"
  - :"gpt-3.5-turbo"
  - :"gpt-3.5-turbo-16k"
  - :"gpt-3.5-turbo-0613"
  - :"gpt-3.5-turbo-1106"
  - :"gpt-3.5-turbo-0125"
  - :"gpt-3.5-turbo-16k-0613"
name: String

The name of the assistant. The maximum length is 256 characters.
reasoning_effort: 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: 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.
- AssistantResponseFormatOption = :auto
  
  auto is the default value
  - :auto
- class ResponseFormatText
  
  Default response format. Used to generate text responses.
  - type: :text
    
    The type of response format being defined. Always text.
    - :text
- class ResponseFormatJSONObject
  
  JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
  - type: :json_object
    
    The type of response format being defined. Always json_object.
    - :json_object
- class ResponseFormatJSONSchema
  
  JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
  - json_schema: JSONSchema{ 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: String
      
      A description of what the response format is for, used by the model to determine how to respond in the format.
    - schema: Hash[Symbol, untyped]
      
      The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
    - strict: bool
      
      Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
  - type: :json_schema
    
    The type of response format being defined. Always json_schema.
    - :json_schema
temperature: Float

What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
tool_resources: ToolResources{ 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: CodeInterpreter{ file_ids}
  - file_ids: Array[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: FileSearch{ vector_store_ids}
  - vector_store_ids: Array[String]
    
    Overrides the vector store attached to this assistant. There can be a maximum of 1 vector store attached to the assistant.
tools: Array[AssistantTool]

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.
- class CodeInterpreterTool
  - type: :code_interpreter
    
    The type of tool being defined: code_interpreter
    - :code_interpreter
- class FileSearchTool
  - type: :file_search
    
    The type of tool being defined: file_search
    - :file_search
  - file_search: FileSearch{ max_num_results, ranking_options}
    
    Overrides for the file search tool.
    - 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.
    - ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
      - ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
- class FunctionTool
  - 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: String
      
      A description of what the function does, used by the model to choose when and how to call the function.
    - parameters: 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: bool
      
      Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
  - type: :function
    
    The type of tool being defined: function
    - :function
top_p: Float

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

We generally recommend altering this or temperature but not both.

Returns

class Assistant

Represents an assistant that can call the model and use tools.
- id: String
  
  The identifier, which can be referenced in API endpoints.
- created_at: Integer
  
  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[AssistantTool]
  
  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.
  - class CodeInterpreterTool
    - type: :code_interpreter
      
      The type of tool being defined: code_interpreter
      - :code_interpreter
  - class FileSearchTool
    - type: :file_search
      
      The type of tool being defined: file_search
      - :file_search
    - file_search: FileSearch{ max_num_results, ranking_options}
      
      Overrides for the file search tool.
      - 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.
      - ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
  - class FunctionTool
    - 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: String
        
        A description of what the function does, used by the model to choose when and how to call the function.
      - parameters: 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: bool
        
        Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
    - type: :function
      
      The type of tool being defined: function
      - :function
- 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.
  - AssistantResponseFormatOption = :auto
    
    auto is the default value
    - :auto
  - class ResponseFormatText
    
    Default response format. Used to generate text responses.
    - type: :text
      
      The type of response format being defined. Always text.
      - :text
  - class ResponseFormatJSONObject
    
    JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
    - type: :json_object
      
      The type of response format being defined. Always json_object.
      - :json_object
  - class ResponseFormatJSONSchema
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: JSONSchema{ 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: String
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: Hash[Symbol, untyped]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: bool
        
        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
    - type: :json_schema
      
      The type of response format being defined. Always json_schema.
      - :json_schema
- temperature: Float
  
  What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
- tool_resources: ToolResources{ 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: CodeInterpreter{ file_ids}
    - 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.
  - file_search: FileSearch{ vector_store_ids}
    - 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.
- top_p: Float
  
  An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
  
  We generally recommend altering this or temperature but not both.

Example

require "openai"

openai = OpenAI::Client.new(api_key: "My API Key")

assistant = openai.beta.assistants.update("assistant_id")

puts(assistant)

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
}

Delete assistant

beta.assistants.delete(assistant_id) -> AssistantDeleted

delete /assistants/{assistant_id}

Delete an assistant.

Parameters

assistant_id: String

Returns

class AssistantDeleted
- id: String
- deleted: bool
- object: :"assistant.deleted"
  - :"assistant.deleted"

Example

require "openai"

openai = OpenAI::Client.new(api_key: "My API Key")

assistant_deleted = openai.beta.assistants.delete("assistant_id")

puts(assistant_deleted)

Response

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

Domain Types

Assistant

class Assistant

Represents an assistant that can call the model and use tools.
- id: String
  
  The identifier, which can be referenced in API endpoints.
- created_at: Integer
  
  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[AssistantTool]
  
  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.
  - class CodeInterpreterTool
    - type: :code_interpreter
      
      The type of tool being defined: code_interpreter
      - :code_interpreter
  - class FileSearchTool
    - type: :file_search
      
      The type of tool being defined: file_search
      - :file_search
    - file_search: FileSearch{ max_num_results, ranking_options}
      
      Overrides for the file search tool.
      - 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.
      - ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
  - class FunctionTool
    - 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: String
        
        A description of what the function does, used by the model to choose when and how to call the function.
      - parameters: 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: bool
        
        Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
    - type: :function
      
      The type of tool being defined: function
      - :function
- 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.
  - AssistantResponseFormatOption = :auto
    
    auto is the default value
    - :auto
  - class ResponseFormatText
    
    Default response format. Used to generate text responses.
    - type: :text
      
      The type of response format being defined. Always text.
      - :text
  - class ResponseFormatJSONObject
    
    JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
    - type: :json_object
      
      The type of response format being defined. Always json_object.
      - :json_object
  - class ResponseFormatJSONSchema
    
    JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
    - json_schema: JSONSchema{ 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: String
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
      - schema: Hash[Symbol, untyped]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
      - strict: bool
        
        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
    - type: :json_schema
      
      The type of response format being defined. Always json_schema.
      - :json_schema
- temperature: Float
  
  What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
- tool_resources: ToolResources{ 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: CodeInterpreter{ file_ids}
    - 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.
  - file_search: FileSearch{ vector_store_ids}
    - 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.
- top_p: Float
  
  An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
  
  We generally recommend altering this or temperature but not both.

Assistant Deleted

class AssistantDeleted
- id: String
- deleted: bool
- object: :"assistant.deleted"
  - :"assistant.deleted"

Assistant Stream Event

AssistantStreamEvent = ThreadCreated{ data, event, enabled} | ThreadRunCreated{ data, event} | ThreadRunQueued{ data, event} | 21 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.
- class ThreadCreated
  
  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: Integer
      
      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: ToolResources{ 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: CodeInterpreter{ file_ids}
        
        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.
      - file_search: FileSearch{ vector_store_ids}
        
        vector_store_ids: Array[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: bool
    
    Whether to enable input audio transcription.
- class ThreadRunCreated
  
  Occurs when a new run is created.
  - data: Run
    
    Represents an execution run on a thread.
    - id: 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: Integer
      
      The Unix timestamp (in seconds) for when the run was cancelled.
    - completed_at: Integer
      
      The Unix timestamp (in seconds) for when the run was completed.
    - created_at: Integer
      
      The Unix timestamp (in seconds) for when the run was created.
    - expires_at: Integer
      
      The Unix timestamp (in seconds) for when the run will expire.
    - failed_at: Integer
      
      The Unix timestamp (in seconds) for when the run failed.
    - incomplete_details: IncompleteDetails{ reason}
      
      Details on why the run is incomplete. Will be null if the run is not incomplete.
      - reason: :max_completion_tokens | :max_prompt_tokens
        
        The reason why the run is incomplete. This will point to which specific token limit was reached over the course of the run.
        
        :max_completion_tokens
        
        :max_prompt_tokens
    - instructions: String
      
      The instructions that the assistant used for this run.
    - last_error: LastError{ code, message}
      
      The last error associated with this run. Will be null if there are no errors.
      - code: :server_error | :rate_limit_exceeded | :invalid_prompt
        
        One of server_error, rate_limit_exceeded, or invalid_prompt.
        
        :server_error
        
        :rate_limit_exceeded
        
        :invalid_prompt
      - message: String
        
        A human-readable description of the error.
    - max_completion_tokens: Integer
      
      The maximum number of completion tokens specified to have been used over the course of the run.
    - max_prompt_tokens: Integer
      
      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: bool
      
      Whether to enable parallel function calling during tool use.
    - required_action: RequiredAction{ submit_tool_outputs, type}
      
      Details on the action required to continue the run. Will be null if no action is required.
      - submit_tool_outputs: SubmitToolOutputs{ tool_calls}
        
        Details on the tool outputs needed for this run to continue.
        
        tool_calls: Array[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: Function{ 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.
      - AssistantResponseFormatOption = :auto
        
        auto is the default value
        
        :auto
      - class ResponseFormatText
        
        Default response format. Used to generate text responses.
        
        type: :text
        
        The type of response format being defined. Always text.
        
        :text
      - class ResponseFormatJSONObject
        
        JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
        
        type: :json_object
        
        The type of response format being defined. Always json_object.
        
        :json_object
      - class ResponseFormatJSONSchema
        
        JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
        
        json_schema: JSONSchema{ 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: String
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
        
        schema: Hash[Symbol, untyped]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
        
        strict: bool
        
        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
        
        type: :json_schema
        
        The type of response format being defined. Always json_schema.
        
        :json_schema
    - started_at: Integer
      
      The Unix timestamp (in seconds) for when the run was started.
    - status: RunStatus
      
      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.
      - Auto = :none | :auto | :required
        
        none means the model will not call any tools and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools before responding to the user.
        
        :none
        
        :auto
        
        :required
      - class AssistantToolChoice
        
        Specifies a tool the model should use. Use to force the model to call a specific tool.
        
        type: :function | :code_interpreter | :file_search
        
        The type of the tool. If type is function, the function name must be set
        
        :function
        
        :code_interpreter
        
        :file_search
        
        function: AssistantToolChoiceFunction
        
        name: String
        
        The name of the function to call.
    - tools: Array[AssistantTool]
      
      The list of tools that the assistant used for this run.
      - class CodeInterpreterTool
        
        type: :code_interpreter
        
        The type of tool being defined: code_interpreter
        
        :code_interpreter
      - class FileSearchTool
        
        type: :file_search
        
        The type of tool being defined: file_search
        
        :file_search
        
        file_search: FileSearch{ max_num_results, ranking_options}
        
        Overrides for the file search tool.
        
        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.
        
        ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
      - class FunctionTool
        
        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: String
        
        A description of what the function does, used by the model to choose when and how to call the function.
        
        parameters: 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: bool
        
        Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
        
        type: :function
        
        The type of tool being defined: function
        
        :function
    - truncation_strategy: TruncationStrategy{ 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 | :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: Integer
        
        The number of most recent messages from the thread when constructing the context for the run.
    - usage: Usage{ 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: Integer
        
        Number of completion tokens used over the course of the run.
      - prompt_tokens: Integer
        
        Number of prompt tokens used over the course of the run.
      - total_tokens: Integer
        
        Total number of tokens used (prompt + completion).
    - temperature: Float
      
      The sampling temperature used for this run. If not set, defaults to 1.
    - top_p: Float
      
      The nucleus sampling value used for this run. If not set, defaults to 1.
  - event: :"thread.run.created"
    - :"thread.run.created"
- class ThreadRunQueued
  
  Occurs when a run moves to a queued status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.queued"
    - :"thread.run.queued"
- class ThreadRunInProgress
  
  Occurs when a run moves to an in_progress status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.in_progress"
    - :"thread.run.in_progress"
- class ThreadRunRequiresAction
  
  Occurs when a run moves to a requires_action status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.requires_action"
    - :"thread.run.requires_action"
- class ThreadRunCompleted
  
  Occurs when a run is completed.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.completed"
    - :"thread.run.completed"
- class ThreadRunIncomplete
  
  Occurs when a run ends with status incomplete.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.incomplete"
    - :"thread.run.incomplete"
- class ThreadRunFailed
  
  Occurs when a run fails.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.failed"
    - :"thread.run.failed"
- class ThreadRunCancelling
  
  Occurs when a run moves to a cancelling status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.cancelling"
    - :"thread.run.cancelling"
- class ThreadRunCancelled
  
  Occurs when a run is cancelled.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.cancelled"
    - :"thread.run.cancelled"
- class ThreadRunExpired
  
  Occurs when a run expires.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.expired"
    - :"thread.run.expired"
- class ThreadRunStepCreated
  
  Occurs when a run step is created.
  - data: RunStep
    
    Represents a step in execution of a run.
    - id: 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: Integer
      
      The Unix timestamp (in seconds) for when the run step was cancelled.
    - completed_at: Integer
      
      The Unix timestamp (in seconds) for when the run step completed.
    - created_at: Integer
      
      The Unix timestamp (in seconds) for when the run step was created.
    - expired_at: Integer
      
      The Unix timestamp (in seconds) for when the run step expired. A step is considered expired if the parent run is expired.
    - failed_at: Integer
      
      The Unix timestamp (in seconds) for when the run step failed.
    - last_error: LastError{ code, message}
      
      The last error associated with this run step. Will be null if there are no errors.
      - code: :server_error | :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 | :cancelled | :failed | 2 more
      
      The status of the run step, which can be either in_progress, cancelled, failed, completed, or expired.
      - :in_progress
      - :cancelled
      - :failed
      - :completed
      - :expired
    - step_details: MessageCreationStepDetails | ToolCallsStepDetails
      
      The details of the run step.
      - class MessageCreationStepDetails
        
        Details of the message creation by the run step.
        
        message_creation: MessageCreation{ 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
      - class ToolCallsStepDetails
        
        Details of the tool call.
        
        tool_calls: Array[ToolCall]
        
        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.
        
        class CodeInterpreterToolCall
        
        Details of the Code Interpreter tool call the run step was involved in.
        
        id: String
        
        The ID of the tool call.
        
        code_interpreter: CodeInterpreter{ input, outputs}
        
        The Code Interpreter tool call definition.
        
        input: String
        
        The input to the Code Interpreter tool call.
        
        outputs: Array[Logs{ logs, type} | Image{ 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.
        
        class Logs
        
        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
        
        class Image
        
        image: Image{ 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
        
        class FileSearchToolCall
        
        id: String
        
        The ID of the tool call object.
        
        file_search: FileSearch{ ranking_options, results}
        
        For now, this is always going to be an empty object.
        
        ranking_options: RankingOptions{ ranker, score_threshold}
        
        The ranking options for the file search.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
        
        score_threshold: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        results: Array[Result{ 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: Float
        
        The score of the result. All values must be a floating point number between 0 and 1.
        
        content: Array[Content{ text, type}]
        
        The content of the result that was found. The content is only included if requested via the include query parameter.
        
        text: String
        
        The text content of the file.
        
        type: :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
        
        class FunctionToolCall
        
        id: String
        
        The ID of the tool call object.
        
        function: Function{ 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 | :tool_calls
      
      The type of run step, which can be either message_creation or tool_calls.
      - :message_creation
      - :tool_calls
    - usage: Usage{ 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: Integer
        
        Number of completion tokens used over the course of the run step.
      - prompt_tokens: Integer
        
        Number of prompt tokens used over the course of the run step.
      - total_tokens: Integer
        
        Total number of tokens used (prompt + completion).
  - event: :"thread.run.step.created"
    - :"thread.run.step.created"
- class ThreadRunStepInProgress
  
  Occurs when a run step moves to an in_progress state.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: :"thread.run.step.in_progress"
    - :"thread.run.step.in_progress"
- class ThreadRunStepDelta
  
  Occurs when parts of a run step are being streamed.
  - data: RunStepDeltaEvent
    
    Represents a run step delta i.e. any changed fields on a run step during streaming.
    - id: String
      
      The identifier of the run step, which can be referenced in API endpoints.
    - delta: RunStepDelta
      
      The delta containing the fields that have changed on the run step.
      - step_details: RunStepDeltaMessageDelta | ToolCallDeltaObject
        
        The details of the run step.
        
        class RunStepDeltaMessageDelta
        
        Details of the message creation by the run step.
        
        type: :message_creation
        
        Always message_creation.
        
        :message_creation
        
        message_creation: MessageCreation{ message_id}
        
        message_id: String
        
        The ID of the message that was created by this run step.
        
        class ToolCallDeltaObject
        
        Details of the tool call.
        
        type: :tool_calls
        
        Always tool_calls.
        
        :tool_calls
        
        tool_calls: Array[ToolCallDelta]
        
        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.
        
        class CodeInterpreterToolCallDelta
        
        Details of the Code Interpreter tool call the run step was involved in.
        
        index: Integer
        
        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: String
        
        The ID of the tool call.
        
        code_interpreter: CodeInterpreter{ input, outputs}
        
        The Code Interpreter tool call definition.
        
        input: String
        
        The input to the Code Interpreter tool call.
        
        outputs: Array[CodeInterpreterLogs | 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.
        
        class CodeInterpreterLogs
        
        Text output from the Code Interpreter tool call as part of a run step.
        
        index: Integer
        
        The index of the output in the outputs array.
        
        type: :logs
        
        Always logs.
        
        :logs
        
        logs: String
        
        The text output from the Code Interpreter tool call.
        
        class CodeInterpreterOutputImage
        
        index: Integer
        
        The index of the output in the outputs array.
        
        type: :image
        
        Always image.
        
        :image
        
        image: Image{ file_id}
        
        file_id: String
        
        The file ID of the image.
        
        class FileSearchToolCallDelta
        
        file_search: untyped
        
        For now, this is always going to be an empty object.
        
        index: Integer
        
        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: String
        
        The ID of the tool call object.
        
        class FunctionToolCallDelta
        
        index: Integer
        
        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: String
        
        The ID of the tool call object.
        
        function: Function{ 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.
    - 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"
- class ThreadRunStepCompleted
  
  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"
- class ThreadRunStepFailed
  
  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"
- class ThreadRunStepCancelled
  
  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"
- class ThreadRunStepExpired
  
  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"
- class ThreadMessageCreated
  
  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[Attachment{ file_id, tools}]
      
      A list of files attached to the message, and the tools they were added to.
      - file_id: String
        
        The ID of the file to attach to the message.
      - tools: Array[CodeInterpreterTool | AssistantToolsFileSearchTypeOnly{ type}]
        
        The tools to add this file to.
        
        class CodeInterpreterTool
        
        class AssistantToolsFileSearchTypeOnly
        
        type: :file_search
        
        The type of tool being defined: file_search
        
        :file_search
    - completed_at: Integer
      
      The Unix timestamp (in seconds) for when the message was completed.
    - content: Array[MessageContent]
      
      The content of the message in array of text and/or images.
      - class ImageFileContentBlock
        
        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: :auto | :low | :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
      - class ImageURLContentBlock
        
        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: :auto | :low | :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
      - class TextContentBlock
        
        The text content that is part of a message.
        
        text: Text
        
        annotations: Array[Annotation]
        
        class FileCitationAnnotation
        
        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: Integer
        
        file_citation: FileCitation{ file_id}
        
        file_id: String
        
        The ID of the specific File the citation is from.
        
        start_index: Integer
        
        text: String
        
        The text in the message content that needs to be replaced.
        
        type: :file_citation
        
        Always file_citation.
        
        :file_citation
        
        class FilePathAnnotation
        
        A URL for the file that's generated when the assistant used the code_interpreter tool to generate a file.
        
        end_index: Integer
        
        file_path: FilePath{ file_id}
        
        file_id: String
        
        The ID of the file that was generated.
        
        start_index: Integer
        
        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
      - class RefusalContentBlock
        
        The refusal content generated by the assistant.
        
        refusal: String
        
        type: :refusal
        
        Always refusal.
        
        :refusal
    - created_at: Integer
      
      The Unix timestamp (in seconds) for when the message was created.
    - incomplete_at: Integer
      
      The Unix timestamp (in seconds) for when the message was marked as incomplete.
    - incomplete_details: IncompleteDetails{ reason}
      
      On an incomplete message, details about why the message is incomplete.
      - reason: :content_filter | :max_tokens | :run_cancelled | 2 more
        
        The reason the message is incomplete.
        
        :content_filter
        
        :max_tokens
        
        :run_cancelled
        
        :run_expired
        
        :run_failed
    - metadata: 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 | :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 | :incomplete | :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"
- class ThreadMessageInProgress
  
  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"
- class ThreadMessageDelta
  
  Occurs when parts of a Message are being streamed.
  - data: MessageDeltaEvent
    
    Represents a message delta i.e. any changed fields on a message during streaming.
    - id: 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: Array[MessageContentDelta]
        
        The content of the message in array of text and/or images.
        
        class ImageFileDeltaBlock
        
        References an image File in the content of a message.
        
        index: Integer
        
        The index of the content part in the message.
        
        type: :image_file
        
        Always image_file.
        
        :image_file
        
        image_file: ImageFileDelta
        
        detail: :auto | :low | :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: 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.
        
        class TextDeltaBlock
        
        The text content that is part of a message.
        
        index: Integer
        
        The index of the content part in the message.
        
        type: :text
        
        Always text.
        
        :text
        
        text: TextDelta
        
        annotations: Array[AnnotationDelta]
        
        class FileCitationDeltaAnnotation
        
        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: Integer
        
        The index of the annotation in the text content part.
        
        type: :file_citation
        
        Always file_citation.
        
        :file_citation
        
        end_index: Integer
        
        file_citation: FileCitation{ file_id, quote}
        
        file_id: String
        
        The ID of the specific File the citation is from.
        
        quote: String
        
        The specific quote in the file.
        
        start_index: Integer
        
        text: String
        
        The text in the message content that needs to be replaced.
        
        class FilePathDeltaAnnotation
        
        A URL for the file that's generated when the assistant used the code_interpreter tool to generate a file.
        
        index: Integer
        
        The index of the annotation in the text content part.
        
        type: :file_path
        
        Always file_path.
        
        :file_path
        
        end_index: Integer
        
        file_path: FilePath{ file_id}
        
        file_id: String
        
        The ID of the file that was generated.
        
        start_index: Integer
        
        text: String
        
        The text in the message content that needs to be replaced.
        
        value: String
        
        The data that makes up the text.
        
        class RefusalDeltaBlock
        
        The refusal content that is part of a message.
        
        index: Integer
        
        The index of the refusal part in the message.
        
        type: :refusal
        
        Always refusal.
        
        :refusal
        
        refusal: String
        
        class ImageURLDeltaBlock
        
        References an image URL in the content of a message.
        
        index: Integer
        
        The index of the content part in the message.
        
        type: :image_url
        
        Always image_url.
        
        :image_url
        
        image_url: ImageURLDelta
        
        detail: :auto | :low | :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: String
        
        The URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.
      - role: :user | :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"
- class ThreadMessageCompleted
  
  Occurs when a message is completed.
  - data: Message
    
    Represents a message within a thread.
  - event: :"thread.message.completed"
    - :"thread.message.completed"
- class ThreadMessageIncomplete
  
  Occurs when a message ends before it is completed.
  - data: Message
    
    Represents a message within a thread.
  - event: :"thread.message.incomplete"
    - :"thread.message.incomplete"
- class ErrorEvent
  
  Occurs when an error occurs. This can happen due to an internal server error or a timeout.
  - data: ErrorObject
    - code: String
    - message: String
    - param: String
    - type: String
  - event: :error
    - :error

Assistant Tool

AssistantTool = CodeInterpreterTool | FileSearchTool | FunctionTool
- class CodeInterpreterTool
  - type: :code_interpreter
    
    The type of tool being defined: code_interpreter
    - :code_interpreter
- class FileSearchTool
  - type: :file_search
    
    The type of tool being defined: file_search
    - :file_search
  - file_search: FileSearch{ max_num_results, ranking_options}
    
    Overrides for the file search tool.
    - 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.
    - ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
      - ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
- class FunctionTool
  - 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: String
      
      A description of what the function does, used by the model to choose when and how to call the function.
    - parameters: 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: bool
      
      Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
  - type: :function
    
    The type of tool being defined: function
    - :function

Code Interpreter Tool

class CodeInterpreterTool
- type: :code_interpreter
  
  The type of tool being defined: code_interpreter
  - :code_interpreter

File Search Tool

class FileSearchTool
- type: :file_search
  
  The type of tool being defined: file_search
  - :file_search
- file_search: FileSearch{ max_num_results, ranking_options}
  
  Overrides for the file search tool.
  - 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.
  - ranking_options: RankingOptions{ 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: Float
      
      The score threshold for the file search. All values must be a floating point number between 0 and 1.
    - ranker: :auto | :default_2024_08_21
      
      The ranker to use for the file search. If not specified will use the auto ranker.
      - :auto
      - :default_2024_08_21

Function Tool

class FunctionTool
- function: FunctionDefinition
  - name: 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: String
    
    A description of what the function does, used by the model to choose when and how to call the function.
  - parameters: 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: bool
    
    Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
- type: :function
  
  The type of tool being defined: function
  - :function

Message Stream Event

MessageStreamEvent = ThreadMessageCreated{ data, event} | ThreadMessageInProgress{ data, event} | ThreadMessageDelta{ data, event} | 2 more

Occurs when a message is created.
- class ThreadMessageCreated
  
  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[Attachment{ file_id, tools}]
      
      A list of files attached to the message, and the tools they were added to.
      - file_id: String
        
        The ID of the file to attach to the message.
      - tools: Array[CodeInterpreterTool | AssistantToolsFileSearchTypeOnly{ type}]
        
        The tools to add this file to.
        
        class CodeInterpreterTool
        
        type: :code_interpreter
        
        The type of tool being defined: code_interpreter
        
        :code_interpreter
        
        class AssistantToolsFileSearchTypeOnly
        
        type: :file_search
        
        The type of tool being defined: file_search
        
        :file_search
    - completed_at: Integer
      
      The Unix timestamp (in seconds) for when the message was completed.
    - content: Array[MessageContent]
      
      The content of the message in array of text and/or images.
      - class ImageFileContentBlock
        
        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: :auto | :low | :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
      - class ImageURLContentBlock
        
        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: :auto | :low | :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
      - class TextContentBlock
        
        The text content that is part of a message.
        
        text: Text
        
        annotations: Array[Annotation]
        
        class FileCitationAnnotation
        
        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: Integer
        
        file_citation: FileCitation{ file_id}
        
        file_id: String
        
        The ID of the specific File the citation is from.
        
        start_index: Integer
        
        text: String
        
        The text in the message content that needs to be replaced.
        
        type: :file_citation
        
        Always file_citation.
        
        :file_citation
        
        class FilePathAnnotation
        
        A URL for the file that's generated when the assistant used the code_interpreter tool to generate a file.
        
        end_index: Integer
        
        file_path: FilePath{ file_id}
        
        file_id: String
        
        The ID of the file that was generated.
        
        start_index: Integer
        
        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
      - class RefusalContentBlock
        
        The refusal content generated by the assistant.
        
        refusal: String
        
        type: :refusal
        
        Always refusal.
        
        :refusal
    - created_at: Integer
      
      The Unix timestamp (in seconds) for when the message was created.
    - incomplete_at: Integer
      
      The Unix timestamp (in seconds) for when the message was marked as incomplete.
    - incomplete_details: IncompleteDetails{ reason}
      
      On an incomplete message, details about why the message is incomplete.
      - reason: :content_filter | :max_tokens | :run_cancelled | 2 more
        
        The reason the message is incomplete.
        
        :content_filter
        
        :max_tokens
        
        :run_cancelled
        
        :run_expired
        
        :run_failed
    - metadata: 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 | :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 | :incomplete | :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"
- class ThreadMessageInProgress
  
  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"
- class ThreadMessageDelta
  
  Occurs when parts of a Message are being streamed.
  - data: MessageDeltaEvent
    
    Represents a message delta i.e. any changed fields on a message during streaming.
    - id: 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: Array[MessageContentDelta]
        
        The content of the message in array of text and/or images.
        
        class ImageFileDeltaBlock
        
        References an image File in the content of a message.
        
        index: Integer
        
        The index of the content part in the message.
        
        type: :image_file
        
        Always image_file.
        
        :image_file
        
        image_file: ImageFileDelta
        
        detail: :auto | :low | :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: 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.
        
        class TextDeltaBlock
        
        The text content that is part of a message.
        
        index: Integer
        
        The index of the content part in the message.
        
        type: :text
        
        Always text.
        
        :text
        
        text: TextDelta
        
        annotations: Array[AnnotationDelta]
        
        class FileCitationDeltaAnnotation
        
        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: Integer
        
        The index of the annotation in the text content part.
        
        type: :file_citation
        
        Always file_citation.
        
        :file_citation
        
        end_index: Integer
        
        file_citation: FileCitation{ file_id, quote}
        
        file_id: String
        
        The ID of the specific File the citation is from.
        
        quote: String
        
        The specific quote in the file.
        
        start_index: Integer
        
        text: String
        
        The text in the message content that needs to be replaced.
        
        class FilePathDeltaAnnotation
        
        A URL for the file that's generated when the assistant used the code_interpreter tool to generate a file.
        
        index: Integer
        
        The index of the annotation in the text content part.
        
        type: :file_path
        
        Always file_path.
        
        :file_path
        
        end_index: Integer
        
        file_path: FilePath{ file_id}
        
        file_id: String
        
        The ID of the file that was generated.
        
        start_index: Integer
        
        text: String
        
        The text in the message content that needs to be replaced.
        
        value: String
        
        The data that makes up the text.
        
        class RefusalDeltaBlock
        
        The refusal content that is part of a message.
        
        index: Integer
        
        The index of the refusal part in the message.
        
        type: :refusal
        
        Always refusal.
        
        :refusal
        
        refusal: String
        
        class ImageURLDeltaBlock
        
        References an image URL in the content of a message.
        
        index: Integer
        
        The index of the content part in the message.
        
        type: :image_url
        
        Always image_url.
        
        :image_url
        
        image_url: ImageURLDelta
        
        detail: :auto | :low | :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: String
        
        The URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.
      - role: :user | :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"
- class ThreadMessageCompleted
  
  Occurs when a message is completed.
  - data: Message
    
    Represents a message within a thread.
  - event: :"thread.message.completed"
    - :"thread.message.completed"
- class ThreadMessageIncomplete
  
  Occurs when a message ends before it is completed.
  - data: Message
    
    Represents a message within a thread.
  - event: :"thread.message.incomplete"
    - :"thread.message.incomplete"

Run Step Stream Event

RunStepStreamEvent = ThreadRunStepCreated{ data, event} | ThreadRunStepInProgress{ data, event} | ThreadRunStepDelta{ data, event} | 4 more

Occurs when a run step is created.
- class ThreadRunStepCreated
  
  Occurs when a run step is created.
  - data: RunStep
    
    Represents a step in execution of a run.
    - id: 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: Integer
      
      The Unix timestamp (in seconds) for when the run step was cancelled.
    - completed_at: Integer
      
      The Unix timestamp (in seconds) for when the run step completed.
    - created_at: Integer
      
      The Unix timestamp (in seconds) for when the run step was created.
    - expired_at: Integer
      
      The Unix timestamp (in seconds) for when the run step expired. A step is considered expired if the parent run is expired.
    - failed_at: Integer
      
      The Unix timestamp (in seconds) for when the run step failed.
    - last_error: LastError{ code, message}
      
      The last error associated with this run step. Will be null if there are no errors.
      - code: :server_error | :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 | :cancelled | :failed | 2 more
      
      The status of the run step, which can be either in_progress, cancelled, failed, completed, or expired.
      - :in_progress
      - :cancelled
      - :failed
      - :completed
      - :expired
    - step_details: MessageCreationStepDetails | ToolCallsStepDetails
      
      The details of the run step.
      - class MessageCreationStepDetails
        
        Details of the message creation by the run step.
        
        message_creation: MessageCreation{ 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
      - class ToolCallsStepDetails
        
        Details of the tool call.
        
        tool_calls: Array[ToolCall]
        
        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.
        
        class CodeInterpreterToolCall
        
        Details of the Code Interpreter tool call the run step was involved in.
        
        id: String
        
        The ID of the tool call.
        
        code_interpreter: CodeInterpreter{ input, outputs}
        
        The Code Interpreter tool call definition.
        
        input: String
        
        The input to the Code Interpreter tool call.
        
        outputs: Array[Logs{ logs, type} | Image{ 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.
        
        class Logs
        
        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
        
        class Image
        
        image: Image{ 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
        
        class FileSearchToolCall
        
        id: String
        
        The ID of the tool call object.
        
        file_search: FileSearch{ ranking_options, results}
        
        For now, this is always going to be an empty object.
        
        ranking_options: RankingOptions{ ranker, score_threshold}
        
        The ranking options for the file search.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
        
        score_threshold: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        results: Array[Result{ 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: Float
        
        The score of the result. All values must be a floating point number between 0 and 1.
        
        content: Array[Content{ text, type}]
        
        The content of the result that was found. The content is only included if requested via the include query parameter.
        
        text: String
        
        The text content of the file.
        
        type: :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
        
        class FunctionToolCall
        
        id: String
        
        The ID of the tool call object.
        
        function: Function{ 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 | :tool_calls
      
      The type of run step, which can be either message_creation or tool_calls.
      - :message_creation
      - :tool_calls
    - usage: Usage{ 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: Integer
        
        Number of completion tokens used over the course of the run step.
      - prompt_tokens: Integer
        
        Number of prompt tokens used over the course of the run step.
      - total_tokens: Integer
        
        Total number of tokens used (prompt + completion).
  - event: :"thread.run.step.created"
    - :"thread.run.step.created"
- class ThreadRunStepInProgress
  
  Occurs when a run step moves to an in_progress state.
  - data: RunStep
    
    Represents a step in execution of a run.
  - event: :"thread.run.step.in_progress"
    - :"thread.run.step.in_progress"
- class ThreadRunStepDelta
  
  Occurs when parts of a run step are being streamed.
  - data: RunStepDeltaEvent
    
    Represents a run step delta i.e. any changed fields on a run step during streaming.
    - id: String
      
      The identifier of the run step, which can be referenced in API endpoints.
    - delta: RunStepDelta
      
      The delta containing the fields that have changed on the run step.
      - step_details: RunStepDeltaMessageDelta | ToolCallDeltaObject
        
        The details of the run step.
        
        class RunStepDeltaMessageDelta
        
        Details of the message creation by the run step.
        
        type: :message_creation
        
        Always message_creation.
        
        :message_creation
        
        message_creation: MessageCreation{ message_id}
        
        message_id: String
        
        The ID of the message that was created by this run step.
        
        class ToolCallDeltaObject
        
        Details of the tool call.
        
        type: :tool_calls
        
        Always tool_calls.
        
        :tool_calls
        
        tool_calls: Array[ToolCallDelta]
        
        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.
        
        class CodeInterpreterToolCallDelta
        
        Details of the Code Interpreter tool call the run step was involved in.
        
        index: Integer
        
        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: String
        
        The ID of the tool call.
        
        code_interpreter: CodeInterpreter{ input, outputs}
        
        The Code Interpreter tool call definition.
        
        input: String
        
        The input to the Code Interpreter tool call.
        
        outputs: Array[CodeInterpreterLogs | 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.
        
        class CodeInterpreterLogs
        
        Text output from the Code Interpreter tool call as part of a run step.
        
        index: Integer
        
        The index of the output in the outputs array.
        
        type: :logs
        
        Always logs.
        
        :logs
        
        logs: String
        
        The text output from the Code Interpreter tool call.
        
        class CodeInterpreterOutputImage
        
        index: Integer
        
        The index of the output in the outputs array.
        
        type: :image
        
        Always image.
        
        :image
        
        image: Image{ file_id}
        
        file_id: String
        
        The file ID of the image.
        
        class FileSearchToolCallDelta
        
        file_search: untyped
        
        For now, this is always going to be an empty object.
        
        index: Integer
        
        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: String
        
        The ID of the tool call object.
        
        class FunctionToolCallDelta
        
        index: Integer
        
        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: String
        
        The ID of the tool call object.
        
        function: Function{ 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.
    - 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"
- class ThreadRunStepCompleted
  
  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"
- class ThreadRunStepFailed
  
  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"
- class ThreadRunStepCancelled
  
  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"
- class ThreadRunStepExpired
  
  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 = ThreadRunCreated{ data, event} | ThreadRunQueued{ data, event} | ThreadRunInProgress{ data, event} | 7 more

Occurs when a new run is created.
- class ThreadRunCreated
  
  Occurs when a new run is created.
  - data: Run
    
    Represents an execution run on a thread.
    - id: 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: Integer
      
      The Unix timestamp (in seconds) for when the run was cancelled.
    - completed_at: Integer
      
      The Unix timestamp (in seconds) for when the run was completed.
    - created_at: Integer
      
      The Unix timestamp (in seconds) for when the run was created.
    - expires_at: Integer
      
      The Unix timestamp (in seconds) for when the run will expire.
    - failed_at: Integer
      
      The Unix timestamp (in seconds) for when the run failed.
    - incomplete_details: IncompleteDetails{ reason}
      
      Details on why the run is incomplete. Will be null if the run is not incomplete.
      - reason: :max_completion_tokens | :max_prompt_tokens
        
        The reason why the run is incomplete. This will point to which specific token limit was reached over the course of the run.
        
        :max_completion_tokens
        
        :max_prompt_tokens
    - instructions: String
      
      The instructions that the assistant used for this run.
    - last_error: LastError{ code, message}
      
      The last error associated with this run. Will be null if there are no errors.
      - code: :server_error | :rate_limit_exceeded | :invalid_prompt
        
        One of server_error, rate_limit_exceeded, or invalid_prompt.
        
        :server_error
        
        :rate_limit_exceeded
        
        :invalid_prompt
      - message: String
        
        A human-readable description of the error.
    - max_completion_tokens: Integer
      
      The maximum number of completion tokens specified to have been used over the course of the run.
    - max_prompt_tokens: Integer
      
      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: bool
      
      Whether to enable parallel function calling during tool use.
    - required_action: RequiredAction{ submit_tool_outputs, type}
      
      Details on the action required to continue the run. Will be null if no action is required.
      - submit_tool_outputs: SubmitToolOutputs{ tool_calls}
        
        Details on the tool outputs needed for this run to continue.
        
        tool_calls: Array[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: Function{ 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.
      - AssistantResponseFormatOption = :auto
        
        auto is the default value
        
        :auto
      - class ResponseFormatText
        
        Default response format. Used to generate text responses.
        
        type: :text
        
        The type of response format being defined. Always text.
        
        :text
      - class ResponseFormatJSONObject
        
        JSON object response format. An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.
        
        type: :json_object
        
        The type of response format being defined. Always json_object.
        
        :json_object
      - class ResponseFormatJSONSchema
        
        JSON Schema response format. Used to generate structured JSON responses. Learn more about Structured Outputs.
        
        json_schema: JSONSchema{ 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: String
        
        A description of what the response format is for, used by the model to determine how to respond in the format.
        
        schema: Hash[Symbol, untyped]
        
        The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas here.
        
        strict: bool
        
        Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true. To learn more, read the Structured Outputs guide.
        
        type: :json_schema
        
        The type of response format being defined. Always json_schema.
        
        :json_schema
    - started_at: Integer
      
      The Unix timestamp (in seconds) for when the run was started.
    - status: RunStatus
      
      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.
      - Auto = :none | :auto | :required
        
        none means the model will not call any tools and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools before responding to the user.
        
        :none
        
        :auto
        
        :required
      - class AssistantToolChoice
        
        Specifies a tool the model should use. Use to force the model to call a specific tool.
        
        type: :function | :code_interpreter | :file_search
        
        The type of the tool. If type is function, the function name must be set
        
        :function
        
        :code_interpreter
        
        :file_search
        
        function: AssistantToolChoiceFunction
        
        name: String
        
        The name of the function to call.
    - tools: Array[AssistantTool]
      
      The list of tools that the assistant used for this run.
      - class CodeInterpreterTool
        
        type: :code_interpreter
        
        The type of tool being defined: code_interpreter
        
        :code_interpreter
      - class FileSearchTool
        
        type: :file_search
        
        The type of tool being defined: file_search
        
        :file_search
        
        file_search: FileSearch{ max_num_results, ranking_options}
        
        Overrides for the file search tool.
        
        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.
        
        ranking_options: RankingOptions{ 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: Float
        
        The score threshold for the file search. All values must be a floating point number between 0 and 1.
        
        ranker: :auto | :default_2024_08_21
        
        The ranker to use for the file search. If not specified will use the auto ranker.
        
        :auto
        
        :default_2024_08_21
      - class FunctionTool
        
        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: String
        
        A description of what the function does, used by the model to choose when and how to call the function.
        
        parameters: 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: bool
        
        Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is true. Learn more about Structured Outputs in the function calling guide.
        
        type: :function
        
        The type of tool being defined: function
        
        :function
    - truncation_strategy: TruncationStrategy{ 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 | :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: Integer
        
        The number of most recent messages from the thread when constructing the context for the run.
    - usage: Usage{ 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: Integer
        
        Number of completion tokens used over the course of the run.
      - prompt_tokens: Integer
        
        Number of prompt tokens used over the course of the run.
      - total_tokens: Integer
        
        Total number of tokens used (prompt + completion).
    - temperature: Float
      
      The sampling temperature used for this run. If not set, defaults to 1.
    - top_p: Float
      
      The nucleus sampling value used for this run. If not set, defaults to 1.
  - event: :"thread.run.created"
    - :"thread.run.created"
- class ThreadRunQueued
  
  Occurs when a run moves to a queued status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.queued"
    - :"thread.run.queued"
- class ThreadRunInProgress
  
  Occurs when a run moves to an in_progress status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.in_progress"
    - :"thread.run.in_progress"
- class ThreadRunRequiresAction
  
  Occurs when a run moves to a requires_action status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.requires_action"
    - :"thread.run.requires_action"
- class ThreadRunCompleted
  
  Occurs when a run is completed.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.completed"
    - :"thread.run.completed"
- class ThreadRunIncomplete
  
  Occurs when a run ends with status incomplete.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.incomplete"
    - :"thread.run.incomplete"
- class ThreadRunFailed
  
  Occurs when a run fails.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.failed"
    - :"thread.run.failed"
- class ThreadRunCancelling
  
  Occurs when a run moves to a cancelling status.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.cancelling"
    - :"thread.run.cancelling"
- class ThreadRunCancelled
  
  Occurs when a run is cancelled.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.cancelled"
    - :"thread.run.cancelled"
- class ThreadRunExpired
  
  Occurs when a run expires.
  - data: Run
    
    Represents an execution run on a thread.
  - event: :"thread.run.expired"
    - :"thread.run.expired"

Thread Stream Event

class ThreadStreamEvent

Occurs when a new thread is created.
- data: Thread
  
  Represents a thread that contains messages.
  - id: String
    
    The identifier, which can be referenced in API endpoints.
  - created_at: Integer
    
    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: ToolResources{ 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: CodeInterpreter{ file_ids}
      - 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.
    - file_search: FileSearch{ vector_store_ids}
      - vector_store_ids: Array[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: bool
  
  Whether to enable input audio transcription.

ruby/resources/beta/subresources/assistants/index.md +2363 −429

6 6

7**get** `/assistants`7**get** `/assistants`

8 8

~~9List assistants~~9Returns a list of assistants.

10 10

11### Parameters11### Parameters

12 12

75 75

76 - `:assistant`76 - `:assistant`

77 77

~~78 - `tools: Array[untyped]`~~78 - `tools: Array[AssistantTool]`

79 79

80 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`.80 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`.

81 81

82 - `class CodeInterpreterTool`

84 - `type: :code_interpreter`

86 The type of tool being defined: `code_interpreter`

88 - `:code_interpreter`

90 - `class FileSearchTool`

92 - `type: :file_search`

94 The type of tool being defined: `file_search`

96 - `:file_search`

98 - `file_search: FileSearch{ max_num_results, ranking_options}`

100 Overrides for the file search tool.

101

102 - `max_num_results: Integer`

103

104 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.

105

106 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.

107

108 - `ranking_options: RankingOptions{ score_threshold, ranker}`

109

110 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.

111

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

113

114 - `score_threshold: Float`

115

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

117

118 - `ranker: :auto | :default_2024_08_21`

119

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

121

122 - `:auto`

123

124 - `:default_2024_08_21`

125

126 - `class FunctionTool`

127

128 - `function: FunctionDefinition`

129

130 - `name: String`

131

132 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.

133

134 - `description: String`

135

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

137

138 - `parameters: FunctionParameters`

139

140 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.

141

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

143

144 - `strict: bool`

145

146 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).

147

148 - `type: :function`

149

150 The type of tool being defined: `function`

151

152 - `:function`

153

82 - `response_format: AssistantResponseFormatOption`154 - `response_format: AssistantResponseFormatOption`

83 155

84 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`.156 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`.

211 "name": "name",283 "name": "name",

212 "object": "assistant",284 "object": "assistant",

213 "tools": [285 "tools": [

214 {}286 {

287 "type": "code_interpreter"

288 }

215 ],289 ],

216 "response_format": "auto",290 "response_format": "auto",

217 "temperature": 1,291 "temperature": 1,

243 317

244**post** `/assistants`318**post** `/assistants`

245 319

246Create assistant320Create an assistant with a model and instructions.

247 321

248### Parameters322### Parameters

249 323

610 Keys are strings with a maximum length of 64 characters. Values are strings684 Keys are strings with a maximum length of 64 characters. Values are strings

611 with a maximum length of 512 characters.685 with a maximum length of 512 characters.

612 686

613- `tools: Array[untyped]`687- `tools: Array[AssistantTool]`

614 688

615 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`.689 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`.

616 690

691 - `class CodeInterpreterTool`

692

693 - `type: :code_interpreter`

694

695 The type of tool being defined: `code_interpreter`

696

697 - `:code_interpreter`

698

699 - `class FileSearchTool`

700

701 - `type: :file_search`

702

703 The type of tool being defined: `file_search`

704

705 - `:file_search`

706

707 - `file_search: FileSearch{ max_num_results, ranking_options}`

708

709 Overrides for the file search tool.

710

711 - `max_num_results: Integer`

712

713 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.

714

715 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.

716

717 - `ranking_options: RankingOptions{ score_threshold, ranker}`

718

719 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.

720

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

722

723 - `score_threshold: Float`

724

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

726

727 - `ranker: :auto | :default_2024_08_21`

728

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

730

731 - `:auto`

732

733 - `:default_2024_08_21`

734

735 - `class FunctionTool`

736

737 - `function: FunctionDefinition`

738

739 - `name: String`

740

741 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.

742

743 - `description: String`

744

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

746

747 - `parameters: FunctionParameters`

748

749 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.

750

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

752

753 - `strict: bool`

754

755 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).

756

757 - `type: :function`

758

759 The type of tool being defined: `function`

760

761 - `:function`

762

617- `top_p: Float`763- `top_p: Float`

618 764

619 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.765 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.

665 811

666 - `:assistant`812 - `:assistant`

667 813

668 - `tools: Array[untyped]`814 - `tools: Array[AssistantTool]`

669 815

670 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`.816 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`.

671 817

818 - `class CodeInterpreterTool`

819

820 - `type: :code_interpreter`

821

822 The type of tool being defined: `code_interpreter`

823

824 - `:code_interpreter`

825

826 - `class FileSearchTool`

827

828 - `type: :file_search`

829

830 The type of tool being defined: `file_search`

831

832 - `:file_search`

833

834 - `file_search: FileSearch{ max_num_results, ranking_options}`

835

836 Overrides for the file search tool.

837

838 - `max_num_results: Integer`

839

840 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.

841

842 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.

843

844 - `ranking_options: RankingOptions{ score_threshold, ranker}`

845

846 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.

847

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

849

850 - `score_threshold: Float`

851

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

853

854 - `ranker: :auto | :default_2024_08_21`

855

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

857

858 - `:auto`

859

860 - `:default_2024_08_21`

861

862 - `class FunctionTool`

863

864 - `function: FunctionDefinition`

865

866 - `name: String`

867

868 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.

869

870 - `description: String`

871

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

873

874 - `parameters: FunctionParameters`

875

876 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.

877

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

879

880 - `strict: bool`

881

882 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).

883

884 - `type: :function`

885

886 The type of tool being defined: `function`

887

888 - `:function`

889

672 - `response_format: AssistantResponseFormatOption`890 - `response_format: AssistantResponseFormatOption`

673 891

674 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`.892 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`.

799 "name": "name",1017 "name": "name",

800 "object": "assistant",1018 "object": "assistant",

801 "tools": [1019 "tools": [

802 {}1020 {

1021 "type": "code_interpreter"

1022 }

803 ],1023 ],

804 "response_format": "auto",1024 "response_format": "auto",

805 "temperature": 1,1025 "temperature": 1,

825 1045

826**get** `/assistants/{assistant_id}`1046**get** `/assistants/{assistant_id}`

827 1047

828Retrieve assistant1048Retrieves an assistant.

829 1049

830### Parameters1050### Parameters

831 1051

876 1096

877 - `:assistant`1097 - `:assistant`

878 1098

879 - `tools: Array[untyped]`1099 - `tools: Array[AssistantTool]`

880 1100

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`.1101 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 1102

883 - `response_format: AssistantResponseFormatOption`1103 - `class CodeInterpreterTool`

884 1104

885 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`.1105 - `type: :code_interpreter`

886 1106

887 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).1107 The type of tool being defined: `code_interpreter`

888 1108

889 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.1109 - `:code_interpreter`

890 1110

891 **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.1111 - `class FileSearchTool`

892 1112

893 - `AssistantResponseFormatOption = :auto`1113 - `type: :file_search`

894 1114

895 `auto` is the default value1115 The type of tool being defined: `file_search`

896 1116

897 - `:auto`1117 - `:file_search`

898 1118

899 - `class ResponseFormatText`1119 - `file_search: FileSearch{ max_num_results, ranking_options}`

900 1120

901 Default response format. Used to generate text responses.1121 Overrides for the file search tool.

902 1122

903 - `type: :text`1123 - `max_num_results: Integer`

904 1124

905 The type of response format being defined. Always `text`.1125 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 1126

907 - `:text`1127 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.

908 1128

909 - `class ResponseFormatJSONObject`1129 - `ranking_options: RankingOptions{ score_threshold, ranker}`

910 1130

911 JSON object response format. An older method of generating JSON responses.1131 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 Using `json_schema` is recommended for models that support it. Note that the

913 model will not generate JSON without a system or user message instructing it

914 to do so.

915 1132

916 - `type: :json_object`1133 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

917 1134

918 The type of response format being defined. Always `json_object`.1135 - `score_threshold: Float`

919 1136

920 - `:json_object`1137 The score threshold for the file search. All values must be a floating point number between 0 and 1.

921 1138

922 - `class ResponseFormatJSONSchema`1139 - `ranker: :auto | :default_2024_08_21`

923 1140

924 JSON Schema response format. Used to generate structured JSON responses.1141 The ranker to use for the file search. If not specified will use the `auto` ranker.

925 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).1142

1143 - `:auto`

1144

1145 - `:default_2024_08_21`

1146

1147 - `class FunctionTool`

1148

1149 - `function: FunctionDefinition`

1150

1151 - `name: String`

1152

1153 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.

1154

1155 - `description: String`

1156

1157 A description of what the function does, used by the model to choose when and how to call the function.

1158

1159 - `parameters: FunctionParameters`

1160

1161 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.

1162

1163 Omitting `parameters` defines a function with an empty parameter list.

1164

1165 - `strict: bool`

1166

1167 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).

1168

1169 - `type: :function`

1170

1171 The type of tool being defined: `function`

1172

1173 - `:function`

1174

1175 - `response_format: AssistantResponseFormatOption`

1176

1177 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`.

1178

1179 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).

1180

1181 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

1182

1183 **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.

1184

1185 - `AssistantResponseFormatOption = :auto`

1186

1187 `auto` is the default value

1188

1189 - `:auto`

1190

1191 - `class ResponseFormatText`

1192

1193 Default response format. Used to generate text responses.

1194

1195 - `type: :text`

1196

1197 The type of response format being defined. Always `text`.

1198

1199 - `:text`

1200

1201 - `class ResponseFormatJSONObject`

1202

1203 JSON object response format. An older method of generating JSON responses.

1204 Using `json_schema` is recommended for models that support it. Note that the

1205 model will not generate JSON without a system or user message instructing it

1206 to do so.

1207

1208 - `type: :json_object`

1209

1210 The type of response format being defined. Always `json_object`.

1211

1212 - `:json_object`

1213

1214 - `class ResponseFormatJSONSchema`

1215

1216 JSON Schema response format. Used to generate structured JSON responses.

1217 Learn more about [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs).

926 1218

927 - `json_schema: JSONSchema{ name, description, schema, strict}`1219 - `json_schema: JSONSchema{ name, description, schema, strict}`

928 1220

1010 "name": "name",1302 "name": "name",

1011 "object": "assistant",1303 "object": "assistant",

1012 "tools": [1304 "tools": [

1013 {}1305 {

1306 "type": "code_interpreter"

1307 }

1014 ],1308 ],

1015 "response_format": "auto",1309 "response_format": "auto",

1016 "temperature": 1,1310 "temperature": 1,

1036 1330

1037**post** `/assistants/{assistant_id}`1331**post** `/assistants/{assistant_id}`

1038 1332

1039Modify assistant1333Modifies an assistant.

1040 1334

1041### Parameters1335### Parameters

1042 1336

1278 1572

1279 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.1573 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.

1280 1574

1281- `tools: Array[untyped]`1575- `tools: Array[AssistantTool]`

1282 1576

1283 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`.1577 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`.

1284 1578

1579 - `class CodeInterpreterTool`

1580

1581 - `type: :code_interpreter`

1582

1583 The type of tool being defined: `code_interpreter`

1584

1585 - `:code_interpreter`

1586

1587 - `class FileSearchTool`

1588

1589 - `type: :file_search`

1590

1591 The type of tool being defined: `file_search`

1592

1593 - `:file_search`

1594

1595 - `file_search: FileSearch{ max_num_results, ranking_options}`

1596

1597 Overrides for the file search tool.

1598

1599 - `max_num_results: Integer`

1600

1601 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.

1602

1603 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.

1604

1605 - `ranking_options: RankingOptions{ score_threshold, ranker}`

1606

1607 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.

1608

1609 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1610

1611 - `score_threshold: Float`

1612

1613 The score threshold for the file search. All values must be a floating point number between 0 and 1.

1614

1615 - `ranker: :auto | :default_2024_08_21`

1616

1617 The ranker to use for the file search. If not specified will use the `auto` ranker.

1618

1619 - `:auto`

1620

1621 - `:default_2024_08_21`

1622

1623 - `class FunctionTool`

1624

1625 - `function: FunctionDefinition`

1626

1627 - `name: String`

1628

1629 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.

1630

1631 - `description: String`

1632

1633 A description of what the function does, used by the model to choose when and how to call the function.

1634

1635 - `parameters: FunctionParameters`

1636

1637 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.

1638

1639 Omitting `parameters` defines a function with an empty parameter list.

1640

1641 - `strict: bool`

1642

1643 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).

1644

1645 - `type: :function`

1646

1647 The type of tool being defined: `function`

1648

1649 - `:function`

1650

1285- `top_p: Float`1651- `top_p: Float`

1286 1652

1287 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.1653 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.

1333 1699

1334 - `:assistant`1700 - `:assistant`

1335 1701

1336 - `tools: Array[untyped]`1702 - `tools: Array[AssistantTool]`

1337 1703

1338 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`.1704 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`.

1339 1705

1706 - `class CodeInterpreterTool`

1707

1708 - `type: :code_interpreter`

1709

1710 The type of tool being defined: `code_interpreter`

1711

1712 - `:code_interpreter`

1713

1714 - `class FileSearchTool`

1715

1716 - `type: :file_search`

1717

1718 The type of tool being defined: `file_search`

1719

1720 - `:file_search`

1721

1722 - `file_search: FileSearch{ max_num_results, ranking_options}`

1723

1724 Overrides for the file search tool.

1725

1726 - `max_num_results: Integer`

1727

1728 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.

1729

1730 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.

1731

1732 - `ranking_options: RankingOptions{ score_threshold, ranker}`

1733

1734 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.

1735

1736 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1737

1738 - `score_threshold: Float`

1739

1740 The score threshold for the file search. All values must be a floating point number between 0 and 1.

1741

1742 - `ranker: :auto | :default_2024_08_21`

1743

1744 The ranker to use for the file search. If not specified will use the `auto` ranker.

1745

1746 - `:auto`

1747

1748 - `:default_2024_08_21`

1749

1750 - `class FunctionTool`

1751

1752 - `function: FunctionDefinition`

1753

1754 - `name: String`

1755

1756 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.

1757

1758 - `description: String`

1759

1760 A description of what the function does, used by the model to choose when and how to call the function.

1761

1762 - `parameters: FunctionParameters`

1763

1764 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.

1765

1766 Omitting `parameters` defines a function with an empty parameter list.

1767

1768 - `strict: bool`

1769

1770 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).

1771

1772 - `type: :function`

1773

1774 The type of tool being defined: `function`

1775

1776 - `:function`

1777

1340 - `response_format: AssistantResponseFormatOption`1778 - `response_format: AssistantResponseFormatOption`

1341 1779

1342 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`.1780 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`.

1467 "name": "name",1905 "name": "name",

1468 "object": "assistant",1906 "object": "assistant",

1469 "tools": [1907 "tools": [

1470 {}1908 {

1909 "type": "code_interpreter"

1910 }

1471 ],1911 ],

1472 "response_format": "auto",1912 "response_format": "auto",

1473 "temperature": 1,1913 "temperature": 1,

1493 1933

1494**delete** `/assistants/{assistant_id}`1934**delete** `/assistants/{assistant_id}`

1495 1935

1496Delete assistant1936Delete an assistant.

1497 1937

1498### Parameters1938### Parameters

1499 1939

1580 2020

1581 - `:assistant`2021 - `:assistant`

1582 2022

1583 - `tools: Array[untyped]`2023 - `tools: Array[AssistantTool]`

1584 2024

1585 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`.2025 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`.

1586 2026

1587 - `response_format: AssistantResponseFormatOption`2027 - `class CodeInterpreterTool`

1588 2028

1589 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`.2029 - `type: :code_interpreter`

1590 2030

1591 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).2031 The type of tool being defined: `code_interpreter`

1592 2032

1593 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.2033 - `:code_interpreter`

1594 2034

1595 **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.2035 - `class FileSearchTool`

1596 2036

1597 - `AssistantResponseFormatOption = :auto`2037 - `type: :file_search`

1598 2038

1599 `auto` is the default value2039 The type of tool being defined: `file_search`

1600 2040

1601 - `:auto`2041 - `:file_search`

1602 2042

1603 - `class ResponseFormatText`2043 - `file_search: FileSearch{ max_num_results, ranking_options}`

1604 2044

1605 Default response format. Used to generate text responses.2045 Overrides for the file search tool.

1606 2046

1607 - `type: :text`2047 - `max_num_results: Integer`

1608 2048

1609 The type of response format being defined. Always `text`.2049 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.

1610 2050

1611 - `:text`2051 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.

1612 2052

1613 - `class ResponseFormatJSONObject`2053 - `ranking_options: RankingOptions{ score_threshold, ranker}`

1614 2054

1615 JSON object response format. An older method of generating JSON responses.2055 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.

1616 Using `json_schema` is recommended for models that support it. Note that the

1617 model will not generate JSON without a system or user message instructing it

1618 to do so.

1619 2056

1620 - `type: :json_object`2057 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

1621 2058

1622 The type of response format being defined. Always `json_object`.2059 - `score_threshold: Float`

2060

2061 The score threshold for the file search. All values must be a floating point number between 0 and 1.

2062

2063 - `ranker: :auto | :default_2024_08_21`

2064

2065 The ranker to use for the file search. If not specified will use the `auto` ranker.

2066

2067 - `:auto`

2068

2069 - `:default_2024_08_21`

2070

2071 - `class FunctionTool`

2072

2073 - `function: FunctionDefinition`

2074

2075 - `name: String`

2076

2077 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.

2078

2079 - `description: String`

2080

2081 A description of what the function does, used by the model to choose when and how to call the function.

2082

2083 - `parameters: FunctionParameters`

2084

2085 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.

2086

2087 Omitting `parameters` defines a function with an empty parameter list.

2088

2089 - `strict: bool`

2090

2091 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).

2092

2093 - `type: :function`

2094

2095 The type of tool being defined: `function`

2096

2097 - `:function`

2098

2099 - `response_format: AssistantResponseFormatOption`

2100

2101 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`.

2102

2103 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).

2104

2105 Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

2106

2107 **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.

2108

2109 - `AssistantResponseFormatOption = :auto`

2110

2111 `auto` is the default value

2112

2113 - `:auto`

2114

2115 - `class ResponseFormatText`

2116

2117 Default response format. Used to generate text responses.

2118

2119 - `type: :text`

2120

2121 The type of response format being defined. Always `text`.

2122

2123 - `:text`

2124

2125 - `class ResponseFormatJSONObject`

2126

2127 JSON object response format. An older method of generating JSON responses.

2128 Using `json_schema` is recommended for models that support it. Note that the

2129 model will not generate JSON without a system or user message instructing it

2130 to do so.

2131

2132 - `type: :json_object`

2133

2134 The type of response format being defined. Always `json_object`.

1623 2135

1624 - `:json_object`2136 - `:json_object`

1625 2137

2000 2512

2001 The Unix timestamp (in seconds) for when the run was started.2513 The Unix timestamp (in seconds) for when the run was started.

2002 2514

2003 - `status: untyped`2515 - `status: RunStatus`

2516

2517 The status of the run, which can be either `queued`, `in_progress`, `requires_action`, `cancelling`, `cancelled`, `failed`, `completed`, `incomplete`, or `expired`.

2518

2519 - `:queued`

2520

2521 - `:in_progress`

2522

2523 - `:requires_action`

2524

2525 - `:cancelling`

2526

2527 - `:cancelled`

2528

2529 - `:failed`

2530

2531 - `:completed`

2532

2533 - `:incomplete`

2534

2535 - `:expired`

2004 2536

2005 - `thread_id: String`2537 - `thread_id: String`

2006 2538

2044 2576

2045 The name of the function to call.2577 The name of the function to call.

2046 2578

2047 - `tools: Array[untyped]`2579 - `tools: Array[AssistantTool]`

2048 2580

2049 The list of tools that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.2581 The list of tools that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.

2050 2582

2583 - `class CodeInterpreterTool`

2584

2585 - `type: :code_interpreter`

2586

2587 The type of tool being defined: `code_interpreter`

2588

2589 - `:code_interpreter`

2590

2591 - `class FileSearchTool`

2592

2593 - `type: :file_search`

2594

2595 The type of tool being defined: `file_search`

2596

2597 - `:file_search`

2598

2599 - `file_search: FileSearch{ max_num_results, ranking_options}`

2600

2601 Overrides for the file search tool.

2602

2603 - `max_num_results: Integer`

2604

2605 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.

2606

2607 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.

2608

2609 - `ranking_options: RankingOptions{ score_threshold, ranker}`

2610

2611 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.

2612

2613 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

2614

2615 - `score_threshold: Float`

2616

2617 The score threshold for the file search. All values must be a floating point number between 0 and 1.

2618

2619 - `ranker: :auto | :default_2024_08_21`

2620

2621 The ranker to use for the file search. If not specified will use the `auto` ranker.

2622

2623 - `:auto`

2624

2625 - `:default_2024_08_21`

2626

2627 - `class FunctionTool`

2628

2629 - `function: FunctionDefinition`

2630

2631 - `name: String`

2632

2633 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.

2634

2635 - `description: String`

2636

2637 A description of what the function does, used by the model to choose when and how to call the function.

2638

2639 - `parameters: FunctionParameters`

2640

2641 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.

2642

2643 Omitting `parameters` defines a function with an empty parameter list.

2644

2645 - `strict: bool`

2646

2647 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).

2648

2649 - `type: :function`

2650

2651 The type of tool being defined: `function`

2652

2653 - `:function`

2654

2051 - `truncation_strategy: TruncationStrategy{ type, last_messages}`2655 - `truncation_strategy: TruncationStrategy{ type, last_messages}`

2052 2656

2053 Controls for how a thread will be truncated prior to the run. Use this to control the initial context window of the run.2657 Controls for how a thread will be truncated prior to the run. Use this to control the initial context window of the run.

2309 2913

2310 Details of the tool call.2914 Details of the tool call.

2311 2915

2312 - `tool_calls: Array[untyped]`2916 - `tool_calls: Array[ToolCall]`

2313 2917

2314 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`.2918 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`.

2315 2919

2316 - `type: :tool_calls`2920 - `class CodeInterpreterToolCall`

~~2317~~

2318 Always `tool_calls`.

2319 2921

2320 - `:tool_calls`2922 Details of the Code Interpreter tool call the run step was involved in.

2321 2923

2322 - `thread_id: String`2924 - `id: String`

2323 2925

2324 The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) that was run.2926 The ID of the tool call.

2325 2927

2326 - `type: :message_creation | :tool_calls`2928 - `code_interpreter: CodeInterpreter{ input, outputs}`

2327 2929

2328 The type of run step, which can be either `message_creation` or `tool_calls`.2930 The Code Interpreter tool call definition.

2329 2931

2330 - `:message_creation`2932 - `input: String`

2331 2933

2332 - `:tool_calls`2934 The input to the Code Interpreter tool call.

2333 2935

2334 - `usage: Usage{ completion_tokens, prompt_tokens, total_tokens}`2936 - `outputs: Array[Logs{ logs, type} | Image{ image, type}]`

2335 2937

2336 Usage statistics related to the run step. This value will be `null` while the run step's status is `in_progress`.2938 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.

2337 2939

2338 - `completion_tokens: Integer`2940 - `class Logs`

2339 2941

2340 Number of completion tokens used over the course of the run step.2942 Text output from the Code Interpreter tool call as part of a run step.

2341 2943

2342 - `prompt_tokens: Integer`2944 - `logs: String`

2343 2945

2344 Number of prompt tokens used over the course of the run step.2946 The text output from the Code Interpreter tool call.

2345 2947

2346 - `total_tokens: Integer`2948 - `type: :logs`

2347 2949

2348 Total number of tokens used (prompt + completion).2950 Always `logs`.

2349 2951

2350 - `event: :"thread.run.step.created"`2952 - `:logs`

2351 2953

2352 - `:"thread.run.step.created"`2954 - `class Image`

2353 2955

2354 - `class ThreadRunStepInProgress`2956 - `image: Image{ file_id}`

2355 2957

2356 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) moves to an `in_progress` state.2958 - `file_id: String`

2357 2959

2358 - `data: RunStep`2960 The [file](https://platform.openai.com/docs/api-reference/files) ID of the image.

2359 2961

2360 Represents a step in execution of a run.2962 - `type: :image`

2361 2963

2362 - `event: :"thread.run.step.in_progress"`2964 Always `image`.

2363 2965

2364 - `:"thread.run.step.in_progress"`2966 - `:image`

2365 2967

2366 - `class ThreadRunStepDelta`2968 - `type: :code_interpreter`

2367 2969

2368 Occurs when parts of a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) are being streamed.2970 The type of tool call. This is always going to be `code_interpreter` for this type of tool call.

2369 2971

2370 - `data: RunStepDeltaEvent`2972 - `:code_interpreter`

2371 2973

2372 Represents a run step delta i.e. any changed fields on a run step during streaming.2974 - `class FileSearchToolCall`

2373 2975

2374 - `id: String`2976 - `id: String`

2375 2977

2376 The identifier of the run step, which can be referenced in API endpoints.2978 The ID of the tool call object.

2377 2979

2378 - `delta: untyped`2980 - `file_search: FileSearch{ ranking_options, results}`

2379 2981

2380 - `object: :"thread.run.step.delta"`2982 For now, this is always going to be an empty object.

2381 2983

2382 The object type, which is always `thread.run.step.delta`.2984 - `ranking_options: RankingOptions{ ranker, score_threshold}`

2383 2985

2384 - `:"thread.run.step.delta"`2986 The ranking options for the file search.

2385 2987

2386 - `event: :"thread.run.step.delta"`2988 - `ranker: :auto | :default_2024_08_21`

2387 2989

2388 - `:"thread.run.step.delta"`2990 The ranker to use for the file search. If not specified will use the `auto` ranker.

2389 2991

2390 - `class ThreadRunStepCompleted`2992 - `:auto`

2391 2993

2392 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is completed.2994 - `:default_2024_08_21`

2393 2995

2394 - `data: RunStep`2996 - `score_threshold: Float`

2395 2997

2396 Represents a step in execution of a run.2998 The score threshold for the file search. All values must be a floating point number between 0 and 1.

2397 2999

2398 - `event: :"thread.run.step.completed"`3000 - `results: Array[Result{ file_id, file_name, score, content}]`

2399 3001

2400 - `:"thread.run.step.completed"`3002 The results of the file search.

2401 3003

2402 - `class ThreadRunStepFailed`3004 - `file_id: String`

2403 3005

2404 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) fails.3006 The ID of the file that result was found in.

2405 3007

2406 - `data: RunStep`3008 - `file_name: String`

2407 3009

2408 Represents a step in execution of a run.3010 The name of the file that result was found in.

2409 3011

2410 - `event: :"thread.run.step.failed"`3012 - `score: Float`

2411 3013

2412 - `:"thread.run.step.failed"`3014 The score of the result. All values must be a floating point number between 0 and 1.

2413 3015

2414 - `class ThreadRunStepCancelled`3016 - `content: Array[Content{ text, type}]`

2415 3017

2416 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is cancelled.3018 The content of the result that was found. The content is only included if requested via the include query parameter.

2417 3019

2418 - `data: RunStep`3020 - `text: String`

2419 3021

2420 Represents a step in execution of a run.3022 The text content of the file.

2421 3023

2422 - `event: :"thread.run.step.cancelled"`3024 - `type: :text`

2423 3025

2424 - `:"thread.run.step.cancelled"`3026 The type of the content.

2425 3027

2426 - `class ThreadRunStepExpired`3028 - `:text`

2427 3029

2428 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) expires.3030 - `type: :file_search`

2429 3031

2430 - `data: RunStep`3032 The type of tool call. This is always going to be `file_search` for this type of tool call.

2431 3033

2432 Represents a step in execution of a run.3034 - `:file_search`

2433 3035

2434 - `event: :"thread.run.step.expired"`3036 - `class FunctionToolCall`

2435 3037

2436 - `:"thread.run.step.expired"`3038 - `id: String`

2437 3039

2438 - `class ThreadMessageCreated`3040 The ID of the tool call object.

2439 3041

2440 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.3042 - `function: Function{ arguments, name, output}`

2441 3043

2442 - `data: Message`3044 The definition of the function that was called.

2443 3045

2444 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).3046 - `arguments: String`

2445 3047

2446 - `id: String`3048 The arguments passed to the function.

2447 3049

2448 The identifier, which can be referenced in API endpoints.3050 - `name: String`

2449 3051

2450 - `assistant_id: String`3052 The name of the function.

2451 3053

2452 If applicable, the ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) that authored this message.3054 - `output: String`

2453 3055

2454 - `attachments: Array[Attachment{ file_id, tools}]`3056 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.

2455 3057

2456 A list of files attached to the message, and the tools they were added to.3058 - `type: :function`

2457 3059

2458 - `file_id: String`3060 The type of tool call. This is always going to be `function` for this type of tool call.

2459 3061

2460 The ID of the file to attach to the message.3062 - `:function`

2461 3063

2462 - `tools: Array[CodeInterpreterTool | AssistantToolsFileSearchTypeOnly{ type}]`3064 - `type: :tool_calls`

2463 3065

2464 The tools to add this file to.3066 Always `tool_calls`.

3067

3068 - `:tool_calls`

3069

3070 - `thread_id: String`

3071

3072 The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) that was run.

3073

3074 - `type: :message_creation | :tool_calls`

3075

3076 The type of run step, which can be either `message_creation` or `tool_calls`.

3077

3078 - `:message_creation`

3079

3080 - `:tool_calls`

3081

3082 - `usage: Usage{ completion_tokens, prompt_tokens, total_tokens}`

3083

3084 Usage statistics related to the run step. This value will be `null` while the run step's status is `in_progress`.

3085

3086 - `completion_tokens: Integer`

3087

3088 Number of completion tokens used over the course of the run step.

3089

3090 - `prompt_tokens: Integer`

3091

3092 Number of prompt tokens used over the course of the run step.

3093

3094 - `total_tokens: Integer`

3095

3096 Total number of tokens used (prompt + completion).

3097

3098 - `event: :"thread.run.step.created"`

3099

3100 - `:"thread.run.step.created"`

3101

3102 - `class ThreadRunStepInProgress`

3103

3104 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) moves to an `in_progress` state.

3105

3106 - `data: RunStep`

3107

3108 Represents a step in execution of a run.

3109

3110 - `event: :"thread.run.step.in_progress"`

3111

3112 - `:"thread.run.step.in_progress"`

3113

3114 - `class ThreadRunStepDelta`

3115

3116 Occurs when parts of a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) are being streamed.

3117

3118 - `data: RunStepDeltaEvent`

3119

3120 Represents a run step delta i.e. any changed fields on a run step during streaming.

3121

3122 - `id: String`

3123

3124 The identifier of the run step, which can be referenced in API endpoints.

3125

3126 - `delta: RunStepDelta`

3127

3128 The delta containing the fields that have changed on the run step.

3129

3130 - `step_details: RunStepDeltaMessageDelta | ToolCallDeltaObject`

3131

3132 The details of the run step.

3133

3134 - `class RunStepDeltaMessageDelta`

3135

3136 Details of the message creation by the run step.

3137

3138 - `type: :message_creation`

3139

3140 Always `message_creation`.

3141

3142 - `:message_creation`

3143

3144 - `message_creation: MessageCreation{ message_id}`

3145

3146 - `message_id: String`

3147

3148 The ID of the message that was created by this run step.

3149

3150 - `class ToolCallDeltaObject`

3151

3152 Details of the tool call.

3153

3154 - `type: :tool_calls`

3155

3156 Always `tool_calls`.

3157

3158 - `:tool_calls`

3159

3160 - `tool_calls: Array[ToolCallDelta]`

3161

3162 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`.

3163

3164 - `class CodeInterpreterToolCallDelta`

3165

3166 Details of the Code Interpreter tool call the run step was involved in.

3167

3168 - `index: Integer`

3169

3170 The index of the tool call in the tool calls array.

3171

3172 - `type: :code_interpreter`

3173

3174 The type of tool call. This is always going to be `code_interpreter` for this type of tool call.

3175

3176 - `:code_interpreter`

3177

3178 - `id: String`

3179

3180 The ID of the tool call.

3181

3182 - `code_interpreter: CodeInterpreter{ input, outputs}`

3183

3184 The Code Interpreter tool call definition.

3185

3186 - `input: String`

3187

3188 The input to the Code Interpreter tool call.

3189

3190 - `outputs: Array[CodeInterpreterLogs | CodeInterpreterOutputImage]`

3191

3192 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.

3193

3194 - `class CodeInterpreterLogs`

3195

3196 Text output from the Code Interpreter tool call as part of a run step.

3197

3198 - `index: Integer`

3199

3200 The index of the output in the outputs array.

3201

3202 - `type: :logs`

3203

3204 Always `logs`.

3205

3206 - `:logs`

3207

3208 - `logs: String`

3209

3210 The text output from the Code Interpreter tool call.

3211

3212 - `class CodeInterpreterOutputImage`

3213

3214 - `index: Integer`

3215

3216 The index of the output in the outputs array.

3217

3218 - `type: :image`

3219

3220 Always `image`.

3221

3222 - `:image`

3223

3224 - `image: Image{ file_id}`

3225

3226 - `file_id: String`

3227

3228 The [file](https://platform.openai.com/docs/api-reference/files) ID of the image.

3229

3230 - `class FileSearchToolCallDelta`

3231

3232 - `file_search: untyped`

3233

3234 For now, this is always going to be an empty object.

3235

3236 - `index: Integer`

3237

3238 The index of the tool call in the tool calls array.

3239

3240 - `type: :file_search`

3241

3242 The type of tool call. This is always going to be `file_search` for this type of tool call.

3243

3244 - `:file_search`

3245

3246 - `id: String`

3247

3248 The ID of the tool call object.

3249

3250 - `class FunctionToolCallDelta`

3251

3252 - `index: Integer`

3253

3254 The index of the tool call in the tool calls array.

3255

3256 - `type: :function`

3257

3258 The type of tool call. This is always going to be `function` for this type of tool call.

3259

3260 - `:function`

3261

3262 - `id: String`

3263

3264 The ID of the tool call object.

3265

3266 - `function: Function{ arguments, name, output}`

3267

3268 The definition of the function that was called.

3269

3270 - `arguments: String`

3271

3272 The arguments passed to the function.

3273

3274 - `name: String`

3275

3276 The name of the function.

3277

3278 - `output: String`

3279

3280 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.

3281

3282 - `object: :"thread.run.step.delta"`

3283

3284 The object type, which is always `thread.run.step.delta`.

3285

3286 - `:"thread.run.step.delta"`

3287

3288 - `event: :"thread.run.step.delta"`

3289

3290 - `:"thread.run.step.delta"`

3291

3292 - `class ThreadRunStepCompleted`

3293

3294 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is completed.

3295

3296 - `data: RunStep`

3297

3298 Represents a step in execution of a run.

3299

3300 - `event: :"thread.run.step.completed"`

3301

3302 - `:"thread.run.step.completed"`

3303

3304 - `class ThreadRunStepFailed`

3305

3306 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) fails.

3307

3308 - `data: RunStep`

3309

3310 Represents a step in execution of a run.

3311

3312 - `event: :"thread.run.step.failed"`

3313

3314 - `:"thread.run.step.failed"`

3315

3316 - `class ThreadRunStepCancelled`

3317

3318 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is cancelled.

3319

3320 - `data: RunStep`

3321

3322 Represents a step in execution of a run.

3323

3324 - `event: :"thread.run.step.cancelled"`

3325

3326 - `:"thread.run.step.cancelled"`

3327

3328 - `class ThreadRunStepExpired`

3329

3330 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) expires.

3331

3332 - `data: RunStep`

3333

3334 Represents a step in execution of a run.

3335

3336 - `event: :"thread.run.step.expired"`

3337

3338 - `:"thread.run.step.expired"`

3339

3340 - `class ThreadMessageCreated`

3341

3342 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.

3343

3344 - `data: Message`

3345

3346 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).

3347

3348 - `id: String`

3349

3350 The identifier, which can be referenced in API endpoints.

3351

3352 - `assistant_id: String`

3353

3354 If applicable, the ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) that authored this message.

3355

3356 - `attachments: Array[Attachment{ file_id, tools}]`

3357

3358 A list of files attached to the message, and the tools they were added to.

3359

3360 - `file_id: String`

3361

3362 The ID of the file to attach to the message.

3363

3364 - `tools: Array[CodeInterpreterTool | AssistantToolsFileSearchTypeOnly{ type}]`

3365

3366 The tools to add this file to.

3367

3368 - `class CodeInterpreterTool`

3369

3370 - `class AssistantToolsFileSearchTypeOnly`

3371

3372 - `type: :file_search`

3373

3374 The type of tool being defined: `file_search`

3375

3376 - `:file_search`

3377

3378 - `completed_at: Integer`

3379

3380 The Unix timestamp (in seconds) for when the message was completed.

3381

3382 - `content: Array[MessageContent]`

3383

3384 The content of the message in array of text and/or images.

3385

3386 - `class ImageFileContentBlock`

3387

3388 References an image [File](https://platform.openai.com/docs/api-reference/files) in the content of a message.

3389

3390 - `image_file: ImageFile`

3391

3392 - `file_id: String`

3393

3394 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.

3395

3396 - `detail: :auto | :low | :high`

3397

3398 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`.

3399

3400 - `:auto`

3401

3402 - `:low`

3403

3404 - `:high`

3405

3406 - `type: :image_file`

3407

3408 Always `image_file`.

3409

3410 - `:image_file`

3411

3412 - `class ImageURLContentBlock`

3413

3414 References an image URL in the content of a message.

3415

3416 - `image_url: ImageURL`

3417

3418 - `url: String`

3419

3420 The external URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.

3421

3422 - `detail: :auto | :low | :high`

3423

3424 Specifies the detail level of the image. `low` uses fewer tokens, you can opt in to high resolution using `high`. Default value is `auto`

3425

3426 - `:auto`

3427

3428 - `:low`

3429

3430 - `:high`

3431

3432 - `type: :image_url`

3433

3434 The type of the content part.

3435

3436 - `:image_url`

3437

3438 - `class TextContentBlock`

3439

3440 The text content that is part of a message.

3441

3442 - `text: Text`

3443

3444 - `annotations: Array[Annotation]`

3445

3446 - `class FileCitationAnnotation`

3447

3448 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.

3449

3450 - `end_index: Integer`

3451

3452 - `file_citation: FileCitation{ file_id}`

3453

3454 - `file_id: String`

3455

3456 The ID of the specific File the citation is from.

3457

3458 - `start_index: Integer`

3459

3460 - `text: String`

3461

3462 The text in the message content that needs to be replaced.

3463

3464 - `type: :file_citation`

3465

3466 Always `file_citation`.

3467

3468 - `:file_citation`

3469

3470 - `class FilePathAnnotation`

3471

3472 A URL for the file that's generated when the assistant used the `code_interpreter` tool to generate a file.

3473

3474 - `end_index: Integer`

3475

3476 - `file_path: FilePath{ file_id}`

3477

3478 - `file_id: String`

3479

3480 The ID of the file that was generated.

3481

3482 - `start_index: Integer`

3483

3484 - `text: String`

3485

3486 The text in the message content that needs to be replaced.

3487

3488 - `type: :file_path`

3489

3490 Always `file_path`.

3491

3492 - `:file_path`

3493

3494 - `value: String`

3495

3496 The data that makes up the text.

3497

3498 - `type: :text`

3499

3500 Always `text`.

3501

3502 - `:text`

3503

3504 - `class RefusalContentBlock`

3505

3506 The refusal content generated by the assistant.

3507

3508 - `refusal: String`

3509

3510 - `type: :refusal`

3511

3512 Always `refusal`.

3513

3514 - `:refusal`

3515

3516 - `created_at: Integer`

3517

3518 The Unix timestamp (in seconds) for when the message was created.

3519

3520 - `incomplete_at: Integer`

3521

3522 The Unix timestamp (in seconds) for when the message was marked as incomplete.

3523

3524 - `incomplete_details: IncompleteDetails{ reason}`

3525

3526 On an incomplete message, details about why the message is incomplete.

3527

3528 - `reason: :content_filter | :max_tokens | :run_cancelled | 2 more`

3529

3530 The reason the message is incomplete.

3531

3532 - `:content_filter`

3533

3534 - `:max_tokens`

3535

3536 - `:run_cancelled`

3537

3538 - `:run_expired`

3539

3540 - `:run_failed`

3541

3542 - `metadata: Metadata`

3543

3544 Set of 16 key-value pairs that can be attached to an object. This can be

3545 useful for storing additional information about the object in a structured

3546 format, and querying for objects via API or the dashboard.

3547

3548 Keys are strings with a maximum length of 64 characters. Values are strings

3549 with a maximum length of 512 characters.

3550

3551 - `object: :"thread.message"`

3552

3553 The object type, which is always `thread.message`.

3554

3555 - `:"thread.message"`

3556

3557 - `role: :user | :assistant`

3558

3559 The entity that produced the message. One of `user` or `assistant`.

3560

3561 - `:user`

3562

3563 - `:assistant`

3564

3565 - `run_id: String`

3566

3567 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.

3568

3569 - `status: :in_progress | :incomplete | :completed`

3570

3571 The status of the message, which can be either `in_progress`, `incomplete`, or `completed`.

3572

3573 - `:in_progress`

3574

3575 - `:incomplete`

3576

3577 - `:completed`

3578

3579 - `thread_id: String`

3580

3581 The [thread](https://platform.openai.com/docs/api-reference/threads) ID that this message belongs to.

3582

3583 - `event: :"thread.message.created"`

3584

3585 - `:"thread.message.created"`

3586

3587 - `class ThreadMessageInProgress`

3588

3589 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) moves to an `in_progress` state.

3590

3591 - `data: Message`

3592

3593 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).

3594

3595 - `event: :"thread.message.in_progress"`

3596

3597 - `:"thread.message.in_progress"`

3598

3599 - `class ThreadMessageDelta`

3600

3601 Occurs when parts of a [Message](https://platform.openai.com/docs/api-reference/messages/object) are being streamed.

3602

3603 - `data: MessageDeltaEvent`

3604

3605 Represents a message delta i.e. any changed fields on a message during streaming.

3606

3607 - `id: String`

3608

3609 The identifier of the message, which can be referenced in API endpoints.

3610

3611 - `delta: MessageDelta`

3612

3613 The delta containing the fields that have changed on the Message.

3614

3615 - `content: Array[MessageContentDelta]`

3616

3617 The content of the message in array of text and/or images.

3618

3619 - `class ImageFileDeltaBlock`

3620

3621 References an image [File](https://platform.openai.com/docs/api-reference/files) in the content of a message.

3622

3623 - `index: Integer`

3624

3625 The index of the content part in the message.

3626

3627 - `type: :image_file`

3628

3629 Always `image_file`.

3630

3631 - `:image_file`

3632

3633 - `image_file: ImageFileDelta`

3634

3635 - `detail: :auto | :low | :high`

3636

3637 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`.

3638

3639 - `:auto`

3640

3641 - `:low`

3642

3643 - `:high`

3644

3645 - `file_id: String`

3646

3647 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.

3648

3649 - `class TextDeltaBlock`

3650

3651 The text content that is part of a message.

3652

3653 - `index: Integer`

3654

3655 The index of the content part in the message.

3656

3657 - `type: :text`

3658

3659 Always `text`.

3660

3661 - `:text`

3662

3663 - `text: TextDelta`

3664

3665 - `annotations: Array[AnnotationDelta]`

3666

3667 - `class FileCitationDeltaAnnotation`

3668

3669 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.

3670

3671 - `index: Integer`

3672

3673 The index of the annotation in the text content part.

3674

3675 - `type: :file_citation`

3676

3677 Always `file_citation`.

3678

3679 - `:file_citation`

3680

3681 - `end_index: Integer`

3682

3683 - `file_citation: FileCitation{ file_id, quote}`

3684

3685 - `file_id: String`

3686

3687 The ID of the specific File the citation is from.

3688

3689 - `quote: String`

3690

3691 The specific quote in the file.

3692

3693 - `start_index: Integer`

3694

3695 - `text: String`

3696

3697 The text in the message content that needs to be replaced.

3698

3699 - `class FilePathDeltaAnnotation`

3700

3701 A URL for the file that's generated when the assistant used the `code_interpreter` tool to generate a file.

3702

3703 - `index: Integer`

3704

3705 The index of the annotation in the text content part.

3706

3707 - `type: :file_path`

3708

3709 Always `file_path`.

3710

3711 - `:file_path`

3712

3713 - `end_index: Integer`

3714

3715 - `file_path: FilePath{ file_id}`

3716

3717 - `file_id: String`

3718

3719 The ID of the file that was generated.

3720

3721 - `start_index: Integer`

3722

3723 - `text: String`

3724

3725 The text in the message content that needs to be replaced.

3726

3727 - `value: String`

3728

3729 The data that makes up the text.

3730

3731 - `class RefusalDeltaBlock`

3732

3733 The refusal content that is part of a message.

3734

3735 - `index: Integer`

3736

3737 The index of the refusal part in the message.

3738

3739 - `type: :refusal`

3740

3741 Always `refusal`.

3742

3743 - `:refusal`

3744

3745 - `refusal: String`

3746

3747 - `class ImageURLDeltaBlock`

3748

3749 References an image URL in the content of a message.

3750

3751 - `index: Integer`

3752

3753 The index of the content part in the message.

3754

3755 - `type: :image_url`

3756

3757 Always `image_url`.

3758

3759 - `:image_url`

3760

3761 - `image_url: ImageURLDelta`

3762

3763 - `detail: :auto | :low | :high`

3764

3765 Specifies the detail level of the image. `low` uses fewer tokens, you can opt in to high resolution using `high`.

3766

3767 - `:auto`

3768

3769 - `:low`

3770

3771 - `:high`

3772

3773 - `url: String`

3774

3775 The URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.

3776

3777 - `role: :user | :assistant`

3778

3779 The entity that produced the message. One of `user` or `assistant`.

3780

3781 - `:user`

3782

3783 - `:assistant`

3784

3785 - `object: :"thread.message.delta"`

3786

3787 The object type, which is always `thread.message.delta`.

3788

3789 - `:"thread.message.delta"`

3790

3791 - `event: :"thread.message.delta"`

3792

3793 - `:"thread.message.delta"`

3794

3795 - `class ThreadMessageCompleted`

3796

3797 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is completed.

3798

3799 - `data: Message`

3800

3801 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).

3802

3803 - `event: :"thread.message.completed"`

3804

3805 - `:"thread.message.completed"`

3806

3807 - `class ThreadMessageIncomplete`

3808

3809 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) ends before it is completed.

3810

3811 - `data: Message`

3812

3813 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).

3814

3815 - `event: :"thread.message.incomplete"`

3816

3817 - `:"thread.message.incomplete"`

3818

3819 - `class ErrorEvent`

3820

3821 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.

3822

3823 - `data: ErrorObject`

3824

3825 - `code: String`

3826

3827 - `message: String`

3828

3829 - `param: String`

3830

3831 - `type: String`

3832

3833 - `event: :error`

3834

3835 - `:error`

3836

3837### Assistant Tool

3838

3839- `AssistantTool = CodeInterpreterTool | FileSearchTool | FunctionTool`

3840

3841 - `class CodeInterpreterTool`

3842

3843 - `type: :code_interpreter`

3844

3845 The type of tool being defined: `code_interpreter`

3846

3847 - `:code_interpreter`

3848

3849 - `class FileSearchTool`

3850

3851 - `type: :file_search`

3852

3853 The type of tool being defined: `file_search`

3854

3855 - `:file_search`

3856

3857 - `file_search: FileSearch{ max_num_results, ranking_options}`

3858

3859 Overrides for the file search tool.

3860

3861 - `max_num_results: Integer`

3862

3863 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.

3864

3865 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.

3866

3867 - `ranking_options: RankingOptions{ score_threshold, ranker}`

3868

3869 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.

3870

3871 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

3872

3873 - `score_threshold: Float`

3874

3875 The score threshold for the file search. All values must be a floating point number between 0 and 1.

3876

3877 - `ranker: :auto | :default_2024_08_21`

3878

3879 The ranker to use for the file search. If not specified will use the `auto` ranker.

3880

3881 - `:auto`

3882

3883 - `:default_2024_08_21`

3884

3885 - `class FunctionTool`

3886

3887 - `function: FunctionDefinition`

3888

3889 - `name: String`

3890

3891 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.

3892

3893 - `description: String`

3894

3895 A description of what the function does, used by the model to choose when and how to call the function.

3896

3897 - `parameters: FunctionParameters`

3898

3899 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.

3900

3901 Omitting `parameters` defines a function with an empty parameter list.

3902

3903 - `strict: bool`

3904

3905 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).

3906

3907 - `type: :function`

3908

3909 The type of tool being defined: `function`

3910

3911 - `:function`

3912

3913### Code Interpreter Tool

3914

3915- `class CodeInterpreterTool`

3916

3917 - `type: :code_interpreter`

3918

3919 The type of tool being defined: `code_interpreter`

3920

3921 - `:code_interpreter`

3922

3923### File Search Tool

3924

3925- `class FileSearchTool`

3926

3927 - `type: :file_search`

3928

3929 The type of tool being defined: `file_search`

3930

3931 - `:file_search`

3932

3933 - `file_search: FileSearch{ max_num_results, ranking_options}`

3934

3935 Overrides for the file search tool.

3936

3937 - `max_num_results: Integer`

3938

3939 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.

3940

3941 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.

3942

3943 - `ranking_options: RankingOptions{ score_threshold, ranker}`

3944

3945 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.

3946

3947 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

3948

3949 - `score_threshold: Float`

3950

3951 The score threshold for the file search. All values must be a floating point number between 0 and 1.

3952

3953 - `ranker: :auto | :default_2024_08_21`

3954

3955 The ranker to use for the file search. If not specified will use the `auto` ranker.

3956

3957 - `:auto`

3958

3959 - `:default_2024_08_21`

3960

3961### Function Tool

3962

3963- `class FunctionTool`

3964

3965 - `function: FunctionDefinition`

3966

3967 - `name: String`

3968

3969 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.

3970

3971 - `description: String`

3972

3973 A description of what the function does, used by the model to choose when and how to call the function.

3974

3975 - `parameters: FunctionParameters`

3976

3977 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.

3978

3979 Omitting `parameters` defines a function with an empty parameter list.

3980

3981 - `strict: bool`

3982

3983 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).

3984

3985 - `type: :function`

3986

3987 The type of tool being defined: `function`

3988

3989 - `:function`

3990

3991### Message Stream Event

3992

3993- `MessageStreamEvent = ThreadMessageCreated{ data, event} | ThreadMessageInProgress{ data, event} | ThreadMessageDelta{ data, event} | 2 more`

3994

3995 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.

3996

3997 - `class ThreadMessageCreated`

3998

3999 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.

4000

4001 - `data: Message`

4002

4003 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).

4004

4005 - `id: String`

4006

4007 The identifier, which can be referenced in API endpoints.

4008

4009 - `assistant_id: String`

4010

4011 If applicable, the ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) that authored this message.

4012

4013 - `attachments: Array[Attachment{ file_id, tools}]`

4014

4015 A list of files attached to the message, and the tools they were added to.

4016

4017 - `file_id: String`

4018

4019 The ID of the file to attach to the message.

4020

4021 - `tools: Array[CodeInterpreterTool | AssistantToolsFileSearchTypeOnly{ type}]`

4022

4023 The tools to add this file to.

2465 4024

2466 - `class CodeInterpreterTool`4025 - `class CodeInterpreterTool`

2467 4026

2483 4042

2484 The Unix timestamp (in seconds) for when the message was completed.4043 The Unix timestamp (in seconds) for when the message was completed.

2485 4044

2486 - `content: Array[untyped]`4045 - `content: Array[MessageContent]`

2487 4046

2488 The content of the message in array of text and/or images.4047 The content of the message in array of text and/or images.

2489 4048

4049 - `class ImageFileContentBlock`

4050

4051 References an image [File](https://platform.openai.com/docs/api-reference/files) in the content of a message.

4052

4053 - `image_file: ImageFile`

4054

4055 - `file_id: String`

4056

4057 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.

4058

4059 - `detail: :auto | :low | :high`

4060

4061 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`.

4062

4063 - `:auto`

4064

4065 - `:low`

4066

4067 - `:high`

4068

4069 - `type: :image_file`

4070

4071 Always `image_file`.

4072

4073 - `:image_file`

4074

4075 - `class ImageURLContentBlock`

4076

4077 References an image URL in the content of a message.

4078

4079 - `image_url: ImageURL`

4080

4081 - `url: String`

4082

4083 The external URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.

4084

4085 - `detail: :auto | :low | :high`

4086

4087 Specifies the detail level of the image. `low` uses fewer tokens, you can opt in to high resolution using `high`. Default value is `auto`

4088

4089 - `:auto`

4090

4091 - `:low`

4092

4093 - `:high`

4094

4095 - `type: :image_url`

4096

4097 The type of the content part.

4098

4099 - `:image_url`

4100

4101 - `class TextContentBlock`

4102

4103 The text content that is part of a message.

4104

4105 - `text: Text`

4106

4107 - `annotations: Array[Annotation]`

4108

4109 - `class FileCitationAnnotation`

4110

4111 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.

4112

4113 - `end_index: Integer`

4114

4115 - `file_citation: FileCitation{ file_id}`

4116

4117 - `file_id: String`

4118

4119 The ID of the specific File the citation is from.

4120

4121 - `start_index: Integer`

4122

4123 - `text: String`

4124

4125 The text in the message content that needs to be replaced.

4126

4127 - `type: :file_citation`

4128

4129 Always `file_citation`.

4130

4131 - `:file_citation`

4132

4133 - `class FilePathAnnotation`

4134

4135 A URL for the file that's generated when the assistant used the `code_interpreter` tool to generate a file.

4136

4137 - `end_index: Integer`

4138

4139 - `file_path: FilePath{ file_id}`

4140

4141 - `file_id: String`

4142

4143 The ID of the file that was generated.

4144

4145 - `start_index: Integer`

4146

4147 - `text: String`

4148

4149 The text in the message content that needs to be replaced.

4150

4151 - `type: :file_path`

4152

4153 Always `file_path`.

4154

4155 - `:file_path`

4156

4157 - `value: String`

4158

4159 The data that makes up the text.

4160

4161 - `type: :text`

4162

4163 Always `text`.

4164

4165 - `:text`

4166

4167 - `class RefusalContentBlock`

4168

4169 The refusal content generated by the assistant.

4170

4171 - `refusal: String`

4172

4173 - `type: :refusal`

4174

4175 Always `refusal`.

4176

4177 - `:refusal`

4178

2490 - `created_at: Integer`4179 - `created_at: Integer`

2491 4180

2492 The Unix timestamp (in seconds) for when the message was created.4181 The Unix timestamp (in seconds) for when the message was created.

2513 4202

2514 - `:run_failed`4203 - `:run_failed`

2515 4204

2516 - `metadata: Metadata`4205 - `metadata: Metadata`

4206

4207 Set of 16 key-value pairs that can be attached to an object. This can be

4208 useful for storing additional information about the object in a structured

4209 format, and querying for objects via API or the dashboard.

4210

4211 Keys are strings with a maximum length of 64 characters. Values are strings

4212 with a maximum length of 512 characters.

4213

4214 - `object: :"thread.message"`

4215

4216 The object type, which is always `thread.message`.

4217

4218 - `:"thread.message"`

4219

4220 - `role: :user | :assistant`

4221

4222 The entity that produced the message. One of `user` or `assistant`.

4223

4224 - `:user`

4225

4226 - `:assistant`

4227

4228 - `run_id: String`

4229

4230 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.

4231

4232 - `status: :in_progress | :incomplete | :completed`

4233

4234 The status of the message, which can be either `in_progress`, `incomplete`, or `completed`.

4235

4236 - `:in_progress`

4237

4238 - `:incomplete`

4239

4240 - `:completed`

4241

4242 - `thread_id: String`

4243

4244 The [thread](https://platform.openai.com/docs/api-reference/threads) ID that this message belongs to.

4245

4246 - `event: :"thread.message.created"`

4247

4248 - `:"thread.message.created"`

4249

4250 - `class ThreadMessageInProgress`

4251

4252 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) moves to an `in_progress` state.

4253

4254 - `data: Message`

4255

4256 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).

4257

4258 - `event: :"thread.message.in_progress"`

4259

4260 - `:"thread.message.in_progress"`

4261

4262 - `class ThreadMessageDelta`

4263

4264 Occurs when parts of a [Message](https://platform.openai.com/docs/api-reference/messages/object) are being streamed.

4265

4266 - `data: MessageDeltaEvent`

4267

4268 Represents a message delta i.e. any changed fields on a message during streaming.

4269

4270 - `id: String`

4271

4272 The identifier of the message, which can be referenced in API endpoints.

4273

4274 - `delta: MessageDelta`

4275

4276 The delta containing the fields that have changed on the Message.

4277

4278 - `content: Array[MessageContentDelta]`

4279

4280 The content of the message in array of text and/or images.

4281

4282 - `class ImageFileDeltaBlock`

4283

4284 References an image [File](https://platform.openai.com/docs/api-reference/files) in the content of a message.

4285

4286 - `index: Integer`

4287

4288 The index of the content part in the message.

4289

4290 - `type: :image_file`

4291

4292 Always `image_file`.

4293

4294 - `:image_file`

4295

4296 - `image_file: ImageFileDelta`

4297

4298 - `detail: :auto | :low | :high`

4299

4300 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`.

4301

4302 - `:auto`

4303

4304 - `:low`

4305

4306 - `:high`

4307

4308 - `file_id: String`

4309

4310 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.

4311

4312 - `class TextDeltaBlock`

4313

4314 The text content that is part of a message.

4315

4316 - `index: Integer`

4317

4318 The index of the content part in the message.

4319

4320 - `type: :text`

4321

4322 Always `text`.

4323

4324 - `:text`

4325

4326 - `text: TextDelta`

4327

4328 - `annotations: Array[AnnotationDelta]`

4329

4330 - `class FileCitationDeltaAnnotation`

4331

4332 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.

4333

4334 - `index: Integer`

4335

4336 The index of the annotation in the text content part.

4337

4338 - `type: :file_citation`

4339

4340 Always `file_citation`.

4341

4342 - `:file_citation`

4343

4344 - `end_index: Integer`

4345

4346 - `file_citation: FileCitation{ file_id, quote}`

4347

4348 - `file_id: String`

4349

4350 The ID of the specific File the citation is from.

4351

4352 - `quote: String`

4353

4354 The specific quote in the file.

4355

4356 - `start_index: Integer`

4357

4358 - `text: String`

4359

4360 The text in the message content that needs to be replaced.

4361

4362 - `class FilePathDeltaAnnotation`

4363

4364 A URL for the file that's generated when the assistant used the `code_interpreter` tool to generate a file.

4365

4366 - `index: Integer`

2517 4367

2518 Set of 16 key-value pairs that can be attached to an object. This can be4368 The index of the annotation in the text content part.

2519 useful for storing additional information about the object in a structured

2520 format, and querying for objects via API or the dashboard.

2521 4369

2522 Keys are strings with a maximum length of 64 characters. Values are strings4370 - `type: :file_path`

2523 with a maximum length of 512 characters.

2524 4371

2525 - `object: :"thread.message"`4372 Always `file_path`.

2526 4373

2527 The object type, which is always `thread.message`.4374 - `:file_path`

2528 4375

2529 - `:"thread.message"`4376 - `end_index: Integer`

2530 4377

2531 - `role: :user | :assistant`4378 - `file_path: FilePath{ file_id}`

2532 4379

2533 The entity that produced the message. One of `user` or `assistant`.4380 - `file_id: String`

2534 4381

2535 - `:user`4382 The ID of the file that was generated.

2536 4383

2537 - `:assistant`4384 - `start_index: Integer`

2538 4385

2539 - `run_id: String`4386 - `text: String`

2540 4387

2541 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.4388 The text in the message content that needs to be replaced.

2542 4389

2543 - `status: :in_progress | :incomplete | :completed`4390 - `value: String`

2544 4391

2545 The status of the message, which can be either `in_progress`, `incomplete`, or `completed`.4392 The data that makes up the text.

2546 4393

2547 - `:in_progress`4394 - `class RefusalDeltaBlock`

2548 4395

2549 - `:incomplete`4396 The refusal content that is part of a message.

2550 4397

2551 - `:completed`4398 - `index: Integer`

2552 4399

2553 - `thread_id: String`4400 The index of the refusal part in the message.

2554 4401

2555 The [thread](https://platform.openai.com/docs/api-reference/threads) ID that this message belongs to.4402 - `type: :refusal`

2556 4403

2557 - `event: :"thread.message.created"`4404 Always `refusal`.

2558 4405

2559 - `:"thread.message.created"`4406 - `:refusal`

2560 4407

2561 - `class ThreadMessageInProgress`4408 - `refusal: String`

2562 4409

2563 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) moves to an `in_progress` state.4410 - `class ImageURLDeltaBlock`

2564 4411

2565 - `data: Message`4412 References an image URL in the content of a message.

2566 4413

2567 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4414 - `index: Integer`

2568 4415

2569 - `event: :"thread.message.in_progress"`4416 The index of the content part in the message.

2570 4417

2571 - `:"thread.message.in_progress"`4418 - `type: :image_url`

2572 4419

2573 - `class ThreadMessageDelta`4420 Always `image_url`.

2574 4421

2575 Occurs when parts of a [Message](https://platform.openai.com/docs/api-reference/messages/object) are being streamed.4422 - `:image_url`

2576 4423

2577 - `data: MessageDeltaEvent`4424 - `image_url: ImageURLDelta`

2578 4425

2579 Represents a message delta i.e. any changed fields on a message during streaming.4426 - `detail: :auto | :low | :high`

2580 4427

2581 - `id: String`4428 Specifies the detail level of the image. `low` uses fewer tokens, you can opt in to high resolution using `high`.

2582 4429

2583 The identifier of the message, which can be referenced in API endpoints.4430 - `:auto`

2584 4431

2585 - `delta: MessageDelta`4432 - `:low`

2586 4433

2587 The delta containing the fields that have changed on the Message.4434 - `:high`

2588 4435

2589 - `content: Array[untyped]`4436 - `url: String`

2590 4437

2591 The content of the message in array of text and/or images.4438 The URL of the image, must be a supported image types: jpeg, jpg, png, gif, webp.

2592 4439

2593 - `role: :user | :assistant`4440 - `role: :user | :assistant`

2594 4441

2632 4479

2633 - `:"thread.message.incomplete"`4480 - `:"thread.message.incomplete"`

2634 4481

2635 - `class ErrorEvent`4482### Run Step Stream Event

~~2636~~

2637 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.

~~2638~~

2639 - `data: ErrorObject`

~~2640~~

2641 - `code: String`

~~2642~~

2643 - `message: String`

2644 4483

2645 - `param: String`4484- `RunStepStreamEvent = ThreadRunStepCreated{ data, event} | ThreadRunStepInProgress{ data, event} | ThreadRunStepDelta{ data, event} | 4 more`

2646 4485

2647 - `type: String`4486 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is created.

2648 4487

2649 - `event: :error`4488 - `class ThreadRunStepCreated`

2650 4489

2651 - `:error`4490 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is created.

2652 4491

2653### Assistant Tool4492 - `data: RunStep`

2654 4493

2655- `AssistantTool = untyped`4494 Represents a step in execution of a run.

2656 4495

2657### Code Interpreter Tool4496 - `id: String`

2658 4497

2659- `class CodeInterpreterTool`4498 The identifier of the run step, which can be referenced in API endpoints.

2660 4499

2661 - `type: :code_interpreter`4500 - `assistant_id: String`

2662 4501

2663 The type of tool being defined: `code_interpreter`4502 The ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) associated with the run step.

2664 4503

2665 - `:code_interpreter`4504 - `cancelled_at: Integer`

2666 4505

2667### File Search Tool4506 The Unix timestamp (in seconds) for when the run step was cancelled.

2668 4507

2669- `class FileSearchTool`4508 - `completed_at: Integer`

2670 4509

2671 - `type: :file_search`4510 The Unix timestamp (in seconds) for when the run step completed.

2672 4511

2673 The type of tool being defined: `file_search`4512 - `created_at: Integer`

2674 4513

2675 - `:file_search`4514 The Unix timestamp (in seconds) for when the run step was created.

2676 4515

2677 - `file_search: FileSearch{ max_num_results, ranking_options}`4516 - `expired_at: Integer`

2678 4517

2679 Overrides for the file search tool.4518 The Unix timestamp (in seconds) for when the run step expired. A step is considered expired if the parent run is expired.

2680 4519

2681 - `max_num_results: Integer`4520 - `failed_at: Integer`

2682 4521

2683 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.4522 The Unix timestamp (in seconds) for when the run step failed.

2684 4523

2685 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.4524 - `last_error: LastError{ code, message}`

2686 4525

2687 - `ranking_options: RankingOptions{ score_threshold, ranker}`4526 The last error associated with this run step. Will be `null` if there are no errors.

2688 4527

2689 The ranking options for the file search. If not specified, the file search tool will use the `auto` ranker and a score_threshold of 0.4528 - `code: :server_error | :rate_limit_exceeded`

2690 4529

2691 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.4530 One of `server_error` or `rate_limit_exceeded`.

2692 4531

2693 - `score_threshold: Float`4532 - `:server_error`

2694 4533

2695 The score threshold for the file search. All values must be a floating point number between 0 and 1.4534 - `:rate_limit_exceeded`

2696 4535

2697 - `ranker: :auto | :default_2024_08_21`4536 - `message: String`

2698 4537

2699 The ranker to use for the file search. If not specified will use the `auto` ranker.4538 A human-readable description of the error.

2700 4539

2701 - `:auto`4540 - `metadata: Metadata`

2702 4541

2703 - `:default_2024_08_21`4542 Set of 16 key-value pairs that can be attached to an object. This can be

4543 useful for storing additional information about the object in a structured

4544 format, and querying for objects via API or the dashboard.

2704 4545

2705### Function Tool4546 Keys are strings with a maximum length of 64 characters. Values are strings

4547 with a maximum length of 512 characters.

2706 4548

2707- `class FunctionTool`4549 - `object: :"thread.run.step"`

2708 4550

2709 - `function: FunctionDefinition`4551 The object type, which is always `thread.run.step`.

2710 4552

2711 - `name: String`4553 - `:"thread.run.step"`

2712 4554

2713 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 64.4555 - `run_id: String`

2714 4556

2715 - `description: String`4557 The ID of the [run](https://platform.openai.com/docs/api-reference/runs) that this run step is a part of.

2716 4558

2717 A description of what the function does, used by the model to choose when and how to call the function.4559 - `status: :in_progress | :cancelled | :failed | 2 more`

2718 4560

2719 - `parameters: FunctionParameters`4561 The status of the run step, which can be either `in_progress`, `cancelled`, `failed`, `completed`, or `expired`.

2720 4562

2721 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.4563 - `:in_progress`

2722 4564

2723 Omitting `parameters` defines a function with an empty parameter list.4565 - `:cancelled`

2724 4566

2725 - `strict: bool`4567 - `:failed`

2726 4568

2727 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).4569 - `:completed`

2728 4570

2729 - `type: :function`4571 - `:expired`

2730 4572

2731 The type of tool being defined: `function`4573 - `step_details: MessageCreationStepDetails | ToolCallsStepDetails`

2732 4574

2733 - `:function`4575 The details of the run step.

2734 4576

2735### Message Stream Event4577 - `class MessageCreationStepDetails`

2736 4578

2737- `MessageStreamEvent = ThreadMessageCreated{ data, event} | ThreadMessageInProgress{ data, event} | ThreadMessageDelta{ data, event} | 2 more`4579 Details of the message creation by the run step.

2738 4580

2739 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.4581 - `message_creation: MessageCreation{ message_id}`

2740 4582

2741 - `class ThreadMessageCreated`4583 - `message_id: String`

2742 4584

2743 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is created.4585 The ID of the message that was created by this run step.

2744 4586

2745 - `data: Message`4587 - `type: :message_creation`

2746 4588

2747 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4589 Always `message_creation`.

2748 4590

2749 - `id: String`4591 - `:message_creation`

2750 4592

2751 The identifier, which can be referenced in API endpoints.4593 - `class ToolCallsStepDetails`

2752 4594

2753 - `assistant_id: String`4595 Details of the tool call.

2754 4596

2755 If applicable, the ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) that authored this message.4597 - `tool_calls: Array[ToolCall]`

2756 4598

2757 - `attachments: Array[Attachment{ file_id, tools}]`4599 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`.

2758 4600

2759 A list of files attached to the message, and the tools they were added to.4601 - `class CodeInterpreterToolCall`

2760 4602

2761 - `file_id: String`4603 Details of the Code Interpreter tool call the run step was involved in.

2762 4604

2763 The ID of the file to attach to the message.4605 - `id: String`

2764 4606

2765 - `tools: Array[CodeInterpreterTool | AssistantToolsFileSearchTypeOnly{ type}]`4607 The ID of the tool call.

2766 4608

2767 The tools to add this file to.4609 - `code_interpreter: CodeInterpreter{ input, outputs}`

2768 4610

2769 - `class CodeInterpreterTool`4611 The Code Interpreter tool call definition.

2770 4612

2771 - `type: :code_interpreter`4613 - `input: String`

2772 4614

2773 The type of tool being defined: `code_interpreter`4615 The input to the Code Interpreter tool call.

2774 4616

2775 - `:code_interpreter`4617 - `outputs: Array[Logs{ logs, type} | Image{ image, type}]`

2776 4618

2777 - `class AssistantToolsFileSearchTypeOnly`4619 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.

2778 4620

2779 - `type: :file_search`4621 - `class Logs`

2780 4622

2781 The type of tool being defined: `file_search`4623 Text output from the Code Interpreter tool call as part of a run step.

2782 4624

2783 - `:file_search`4625 - `logs: String`

2784 4626

2785 - `completed_at: Integer`4627 The text output from the Code Interpreter tool call.

2786 4628

2787 The Unix timestamp (in seconds) for when the message was completed.4629 - `type: :logs`

2788 4630

2789 - `content: Array[untyped]`4631 Always `logs`.

2790 4632

2791 The content of the message in array of text and/or images.4633 - `:logs`

2792 4634

2793 - `created_at: Integer`4635 - `class Image`

2794 4636

2795 The Unix timestamp (in seconds) for when the message was created.4637 - `image: Image{ file_id}`

2796 4638

2797 - `incomplete_at: Integer`4639 - `file_id: String`

2798 4640

2799 The Unix timestamp (in seconds) for when the message was marked as incomplete.4641 The [file](https://platform.openai.com/docs/api-reference/files) ID of the image.

2800 4642

2801 - `incomplete_details: IncompleteDetails{ reason}`4643 - `type: :image`

2802 4644

2803 On an incomplete message, details about why the message is incomplete.4645 Always `image`.

2804 4646

2805 - `reason: :content_filter | :max_tokens | :run_cancelled | 2 more`4647 - `:image`

2806 4648

2807 The reason the message is incomplete.4649 - `type: :code_interpreter`

2808 4650

2809 - `:content_filter`4651 The type of tool call. This is always going to be `code_interpreter` for this type of tool call.

2810 4652

2811 - `:max_tokens`4653 - `:code_interpreter`

2812 4654

2813 - `:run_cancelled`4655 - `class FileSearchToolCall`

2814 4656

2815 - `:run_expired`4657 - `id: String`

2816 4658

2817 - `:run_failed`4659 The ID of the tool call object.

2818 4660

2819 - `metadata: Metadata`4661 - `file_search: FileSearch{ ranking_options, results}`

2820 4662

2821 Set of 16 key-value pairs that can be attached to an object. This can be4663 For now, this is always going to be an empty object.

2822 useful for storing additional information about the object in a structured

2823 format, and querying for objects via API or the dashboard.

2824 4664

2825 Keys are strings with a maximum length of 64 characters. Values are strings4665 - `ranking_options: RankingOptions{ ranker, score_threshold}`

2826 with a maximum length of 512 characters.

2827 4666

2828 - `object: :"thread.message"`4667 The ranking options for the file search.

2829 4668

2830 The object type, which is always `thread.message`.4669 - `ranker: :auto | :default_2024_08_21`

2831 4670

2832 - `:"thread.message"`4671 The ranker to use for the file search. If not specified will use the `auto` ranker.

2833 4672

2834 - `role: :user | :assistant`4673 - `:auto`

2835 4674

2836 The entity that produced the message. One of `user` or `assistant`.4675 - `:default_2024_08_21`

2837 4676

2838 - `:user`4677 - `score_threshold: Float`

2839 4678

2840 - `:assistant`4679 The score threshold for the file search. All values must be a floating point number between 0 and 1.

2841 4680

2842 - `run_id: String`4681 - `results: Array[Result{ file_id, file_name, score, content}]`

2843 4682

2844 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.4683 The results of the file search.

2845 4684

2846 - `status: :in_progress | :incomplete | :completed`4685 - `file_id: String`

2847 4686

2848 The status of the message, which can be either `in_progress`, `incomplete`, or `completed`.4687 The ID of the file that result was found in.

2849 4688

2850 - `:in_progress`4689 - `file_name: String`

2851 4690

2852 - `:incomplete`4691 The name of the file that result was found in.

2853 4692

2854 - `:completed`4693 - `score: Float`

2855 4694

2856 - `thread_id: String`4695 The score of the result. All values must be a floating point number between 0 and 1.

2857 4696

2858 The [thread](https://platform.openai.com/docs/api-reference/threads) ID that this message belongs to.4697 - `content: Array[Content{ text, type}]`

2859 4698

2860 - `event: :"thread.message.created"`4699 The content of the result that was found. The content is only included if requested via the include query parameter.

2861 4700

2862 - `:"thread.message.created"`4701 - `text: String`

2863 4702

2864 - `class ThreadMessageInProgress`4703 The text content of the file.

2865 4704

2866 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) moves to an `in_progress` state.4705 - `type: :text`

2867 4706

2868 - `data: Message`4707 The type of the content.

2869 4708

2870 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4709 - `:text`

2871 4710

2872 - `event: :"thread.message.in_progress"`4711 - `type: :file_search`

2873 4712

2874 - `:"thread.message.in_progress"`4713 The type of tool call. This is always going to be `file_search` for this type of tool call.

2875 4714

2876 - `class ThreadMessageDelta`4715 - `:file_search`

2877 4716

2878 Occurs when parts of a [Message](https://platform.openai.com/docs/api-reference/messages/object) are being streamed.4717 - `class FunctionToolCall`

2879 4718

2880 - `data: MessageDeltaEvent`4719 - `id: String`

2881 4720

2882 Represents a message delta i.e. any changed fields on a message during streaming.4721 The ID of the tool call object.

2883 4722

2884 - `id: String`4723 - `function: Function{ arguments, name, output}`

2885 4724

2886 The identifier of the message, which can be referenced in API endpoints.4725 The definition of the function that was called.

2887 4726

2888 - `delta: MessageDelta`4727 - `arguments: String`

2889 4728

2890 The delta containing the fields that have changed on the Message.4729 The arguments passed to the function.

2891 4730

2892 - `content: Array[untyped]`4731 - `name: String`

2893 4732

2894 The content of the message in array of text and/or images.4733 The name of the function.

2895 4734

2896 - `role: :user | :assistant`4735 - `output: String`

2897 4736

2898 The entity that produced the message. One of `user` or `assistant`.4737 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.

2899 4738

2900 - `:user`4739 - `type: :function`

2901 4740

2902 - `:assistant`4741 The type of tool call. This is always going to be `function` for this type of tool call.

2903 4742

2904 - `object: :"thread.message.delta"`4743 - `:function`

2905 4744

2906 The object type, which is always `thread.message.delta`.4745 - `type: :tool_calls`

2907 4746

2908 - `:"thread.message.delta"`4747 Always `tool_calls`.

2909 4748

2910 - `event: :"thread.message.delta"`4749 - `:tool_calls`

2911 4750

2912 - `:"thread.message.delta"`4751 - `thread_id: String`

2913 4752

2914 - `class ThreadMessageCompleted`4753 The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) that was run.

2915 4754

2916 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) is completed.4755 - `type: :message_creation | :tool_calls`

2917 4756

2918 - `data: Message`4757 The type of run step, which can be either `message_creation` or `tool_calls`.

2919 4758

2920 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4759 - `:message_creation`

2921 4760

2922 - `event: :"thread.message.completed"`4761 - `:tool_calls`

2923 4762

2924 - `:"thread.message.completed"`4763 - `usage: Usage{ completion_tokens, prompt_tokens, total_tokens}`

2925 4764

2926 - `class ThreadMessageIncomplete`4765 Usage statistics related to the run step. This value will be `null` while the run step's status is `in_progress`.

2927 4766

2928 Occurs when a [message](https://platform.openai.com/docs/api-reference/messages/object) ends before it is completed.4767 - `completion_tokens: Integer`

2929 4768

2930 - `data: Message`4769 Number of completion tokens used over the course of the run step.

2931 4770

2932 Represents a message within a [thread](https://platform.openai.com/docs/api-reference/threads).4771 - `prompt_tokens: Integer`

2933 4772

2934 - `event: :"thread.message.incomplete"`4773 Number of prompt tokens used over the course of the run step.

2935 4774

2936 - `:"thread.message.incomplete"`4775 - `total_tokens: Integer`

2937 4776

2938### Run Step Stream Event4777 Total number of tokens used (prompt + completion).

2939 4778

2940- `RunStepStreamEvent = ThreadRunStepCreated{ data, event} | ThreadRunStepInProgress{ data, event} | ThreadRunStepDelta{ data, event} | 4 more`4779 - `event: :"thread.run.step.created"`

2941 4780

2942 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is created.4781 - `:"thread.run.step.created"`

2943 4782

2944 - `class ThreadRunStepCreated`4783 - `class ThreadRunStepInProgress`

2945 4784

2946 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) is created.4785 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) moves to an `in_progress` state.

2947 4786

2948 - `data: RunStep`4787 - `data: RunStep`

2949 4788

2950 Represents a step in execution of a run.4789 Represents a step in execution of a run.

2951 4790

2952 - `id: String`4791 - `event: :"thread.run.step.in_progress"`

2953 4792

2954 The identifier of the run step, which can be referenced in API endpoints.4793 - `:"thread.run.step.in_progress"`

2955 4794

2956 - `assistant_id: String`4795 - `class ThreadRunStepDelta`

2957 4796

2958 The ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) associated with the run step.4797 Occurs when parts of a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) are being streamed.

2959 4798

2960 - `cancelled_at: Integer`4799 - `data: RunStepDeltaEvent`

2961 4800

2962 The Unix timestamp (in seconds) for when the run step was cancelled.4801 Represents a run step delta i.e. any changed fields on a run step during streaming.

2963 4802

2964 - `completed_at: Integer`4803 - `id: String`

2965 4804

2966 The Unix timestamp (in seconds) for when the run step completed.4805 The identifier of the run step, which can be referenced in API endpoints.

2967 4806

2968 - `created_at: Integer`4807 - `delta: RunStepDelta`

2969 4808

2970 The Unix timestamp (in seconds) for when the run step was created.4809 The delta containing the fields that have changed on the run step.

2971 4810

2972 - `expired_at: Integer`4811 - `step_details: RunStepDeltaMessageDelta | ToolCallDeltaObject`

2973 4812

2974 The Unix timestamp (in seconds) for when the run step expired. A step is considered expired if the parent run is expired.4813 The details of the run step.

2975 4814

2976 - `failed_at: Integer`4815 - `class RunStepDeltaMessageDelta`

2977 4816

2978 The Unix timestamp (in seconds) for when the run step failed.4817 Details of the message creation by the run step.

2979 4818

2980 - `last_error: LastError{ code, message}`4819 - `type: :message_creation`

2981 4820

2982 The last error associated with this run step. Will be `null` if there are no errors.4821 Always `message_creation`.

2983 4822

2984 - `code: :server_error | :rate_limit_exceeded`4823 - `:message_creation`

2985 4824

2986 One of `server_error` or `rate_limit_exceeded`.4825 - `message_creation: MessageCreation{ message_id}`

2987 4826

2988 - `:server_error`4827 - `message_id: String`

2989 4828

2990 - `:rate_limit_exceeded`4829 The ID of the message that was created by this run step.

2991 4830

2992 - `message: String`4831 - `class ToolCallDeltaObject`

2993 4832

2994 A human-readable description of the error.4833 Details of the tool call.

2995 4834

2996 - `metadata: Metadata`4835 - `type: :tool_calls`

2997 4836

2998 Set of 16 key-value pairs that can be attached to an object. This can be4837 Always `tool_calls`.

2999 useful for storing additional information about the object in a structured

3000 format, and querying for objects via API or the dashboard.

3001 4838

3002 Keys are strings with a maximum length of 64 characters. Values are strings4839 - `:tool_calls`

3003 with a maximum length of 512 characters.

3004 4840

3005 - `object: :"thread.run.step"`4841 - `tool_calls: Array[ToolCallDelta]`

3006 4842

3007 The object type, which is always `thread.run.step`.4843 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`.

3008 4844

3009 - `:"thread.run.step"`4845 - `class CodeInterpreterToolCallDelta`

3010 4846

3011 - `run_id: String`4847 Details of the Code Interpreter tool call the run step was involved in.

3012 4848

3013 The ID of the [run](https://platform.openai.com/docs/api-reference/runs) that this run step is a part of.4849 - `index: Integer`

3014 4850

3015 - `status: :in_progress | :cancelled | :failed | 2 more`4851 The index of the tool call in the tool calls array.

3016 4852

3017 The status of the run step, which can be either `in_progress`, `cancelled`, `failed`, `completed`, or `expired`.4853 - `type: :code_interpreter`

3018 4854

3019 - `:in_progress`4855 The type of tool call. This is always going to be `code_interpreter` for this type of tool call.

3020 4856

3021 - `:cancelled`4857 - `:code_interpreter`

3022 4858

3023 - `:failed`4859 - `id: String`

3024 4860

3025 - `:completed`4861 The ID of the tool call.

3026 4862

3027 - `:expired`4863 - `code_interpreter: CodeInterpreter{ input, outputs}`

3028 4864

3029 - `step_details: MessageCreationStepDetails | ToolCallsStepDetails`4865 The Code Interpreter tool call definition.

3030 4866

3031 The details of the run step.4867 - `input: String`

3032 4868

3033 - `class MessageCreationStepDetails`4869 The input to the Code Interpreter tool call.

3034 4870

3035 Details of the message creation by the run step.4871 - `outputs: Array[CodeInterpreterLogs | CodeInterpreterOutputImage]`

3036 4872

3037 - `message_creation: MessageCreation{ message_id}`4873 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.

3038 4874

3039 - `message_id: String`4875 - `class CodeInterpreterLogs`

3040 4876

3041 The ID of the message that was created by this run step.4877 Text output from the Code Interpreter tool call as part of a run step.

3042 4878

3043 - `type: :message_creation`4879 - `index: Integer`

3044 4880

3045 Always `message_creation`.4881 The index of the output in the outputs array.

3046 4882

3047 - `:message_creation`4883 - `type: :logs`

3048 4884

3049 - `class ToolCallsStepDetails`4885 Always `logs`.

3050 4886

3051 Details of the tool call.4887 - `:logs`

3052 4888

3053 - `tool_calls: Array[untyped]`4889 - `logs: String`

3054 4890

3055 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`.4891 The text output from the Code Interpreter tool call.

3056 4892

3057 - `type: :tool_calls`4893 - `class CodeInterpreterOutputImage`

3058 4894

3059 Always `tool_calls`.4895 - `index: Integer`

3060 4896

3061 - `:tool_calls`4897 The index of the output in the outputs array.

3062 4898

3063 - `thread_id: String`4899 - `type: :image`

3064 4900

3065 The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) that was run.4901 Always `image`.

3066 4902

3067 - `type: :message_creation | :tool_calls`4903 - `:image`

3068 4904

3069 The type of run step, which can be either `message_creation` or `tool_calls`.4905 - `image: Image{ file_id}`

3070 4906

3071 - `:message_creation`4907 - `file_id: String`

3072 4908

3073 - `:tool_calls`4909 The [file](https://platform.openai.com/docs/api-reference/files) ID of the image.

3074 4910

3075 - `usage: Usage{ completion_tokens, prompt_tokens, total_tokens}`4911 - `class FileSearchToolCallDelta`

3076 4912

3077 Usage statistics related to the run step. This value will be `null` while the run step's status is `in_progress`.4913 - `file_search: untyped`

3078 4914

3079 - `completion_tokens: Integer`4915 For now, this is always going to be an empty object.

3080 4916

3081 Number of completion tokens used over the course of the run step.4917 - `index: Integer`

3082 4918

3083 - `prompt_tokens: Integer`4919 The index of the tool call in the tool calls array.

3084 4920

3085 Number of prompt tokens used over the course of the run step.4921 - `type: :file_search`

3086 4922

3087 - `total_tokens: Integer`4923 The type of tool call. This is always going to be `file_search` for this type of tool call.

3088 4924

3089 Total number of tokens used (prompt + completion).4925 - `:file_search`

3090 4926

3091 - `event: :"thread.run.step.created"`4927 - `id: String`

3092 4928

3093 - `:"thread.run.step.created"`4929 The ID of the tool call object.

3094 4930

3095 - `class ThreadRunStepInProgress`4931 - `class FunctionToolCallDelta`

3096 4932

3097 Occurs when a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) moves to an `in_progress` state.4933 - `index: Integer`

3098 4934

3099 - `data: RunStep`4935 The index of the tool call in the tool calls array.

3100 4936

3101 Represents a step in execution of a run.4937 - `type: :function`

3102 4938

3103 - `event: :"thread.run.step.in_progress"`4939 The type of tool call. This is always going to be `function` for this type of tool call.

3104 4940

3105 - `:"thread.run.step.in_progress"`4941 - `:function`

3106 4942

3107 - `class ThreadRunStepDelta`4943 - `id: String`

3108 4944

3109 Occurs when parts of a [run step](https://platform.openai.com/docs/api-reference/run-steps/step-object) are being streamed.4945 The ID of the tool call object.

3110 4946

3111 - `data: RunStepDeltaEvent`4947 - `function: Function{ arguments, name, output}`

3112 4948

3113 Represents a run step delta i.e. any changed fields on a run step during streaming.4949 The definition of the function that was called.

3114 4950

3115 - `id: String`4951 - `arguments: String`

3116 4952

3117 The identifier of the run step, which can be referenced in API endpoints.4953 The arguments passed to the function.

4954

4955 - `name: String`

4956

4957 The name of the function.

4958

4959 - `output: String`

3118 4960

3119 - `delta: untyped`4961 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.

3120 4962

3121 - `object: :"thread.run.step.delta"`4963 - `object: :"thread.run.step.delta"`

3122 4964

3404 5246

3405 The Unix timestamp (in seconds) for when the run was started.5247 The Unix timestamp (in seconds) for when the run was started.

3406 5248

3407 - `status: untyped`5249 - `status: RunStatus`

5250

5251 The status of the run, which can be either `queued`, `in_progress`, `requires_action`, `cancelling`, `cancelled`, `failed`, `completed`, `incomplete`, or `expired`.

5252

5253 - `:queued`

5254

5255 - `:in_progress`

5256

5257 - `:requires_action`

5258

5259 - `:cancelling`

5260

5261 - `:cancelled`

5262

5263 - `:failed`

5264

5265 - `:completed`

5266

5267 - `:incomplete`

5268

5269 - `:expired`

3408 5270

3409 - `thread_id: String`5271 - `thread_id: String`

3410 5272

3448 5310

3449 The name of the function to call.5311 The name of the function to call.

3450 5312

3451 - `tools: Array[untyped]`5313 - `tools: Array[AssistantTool]`

3452 5314

3453 The list of tools that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.5315 The list of tools that the [assistant](https://platform.openai.com/docs/api-reference/assistants) used for this run.

3454 5316

5317 - `class CodeInterpreterTool`

5318

5319 - `type: :code_interpreter`

5320

5321 The type of tool being defined: `code_interpreter`

5322

5323 - `:code_interpreter`

5324

5325 - `class FileSearchTool`

5326

5327 - `type: :file_search`

5328

5329 The type of tool being defined: `file_search`

5330

5331 - `:file_search`

5332

5333 - `file_search: FileSearch{ max_num_results, ranking_options}`

5334

5335 Overrides for the file search tool.

5336

5337 - `max_num_results: Integer`

5338

5339 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.

5340

5341 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.

5342

5343 - `ranking_options: RankingOptions{ score_threshold, ranker}`

5344

5345 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.

5346

5347 See the [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings) for more information.

5348

5349 - `score_threshold: Float`

5350

5351 The score threshold for the file search. All values must be a floating point number between 0 and 1.

5352

5353 - `ranker: :auto | :default_2024_08_21`

5354

5355 The ranker to use for the file search. If not specified will use the `auto` ranker.

5356

5357 - `:auto`

5358

5359 - `:default_2024_08_21`

5360

5361 - `class FunctionTool`

5362

5363 - `function: FunctionDefinition`

5364

5365 - `name: String`

5366

5367 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.

5368

5369 - `description: String`

5370

5371 A description of what the function does, used by the model to choose when and how to call the function.

5372

5373 - `parameters: FunctionParameters`

5374

5375 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.

5376

5377 Omitting `parameters` defines a function with an empty parameter list.

5378

5379 - `strict: bool`

5380

5381 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).

5382

5383 - `type: :function`

5384

5385 The type of tool being defined: `function`

5386

5387 - `:function`

5388

3455 - `truncation_strategy: TruncationStrategy{ type, last_messages}`5389 - `truncation_strategy: TruncationStrategy{ type, last_messages}`

3456 5390

3457 Controls for how a thread will be truncated prior to the run. Use this to control the initial context window of the run.5391 Controls for how a thread will be truncated prior to the run. Use this to control the initial context window of the run.

ruby/resources/beta/subresources/assistants/index.md 2026-07-10 23:02 UTC to 2026-07-12 06:58 UTC

Assistants

List assistants

Parameters

Returns

Example

Response

Create assistant

Parameters

Returns

Example

Response

Retrieve assistant

Parameters

Returns

Example

Response

Modify assistant

Parameters

Returns

Example

Response

Delete assistant

Parameters

Returns

Example

Response

Domain Types

Assistant

Assistant Deleted

Assistant Stream Event

Assistant Tool

Code Interpreter Tool

File Search Tool

Function Tool

Message Stream Event

Run Step Stream Event

Run Stream Event

Thread Stream Event