ChatGPT · Schema

CreateResponseRequest

AgentsAIChatGPTEmbeddingsFine-TuningGPT-4GPT-5Language ModelOpenAIRealtime

Properties

Name	Type	Description
model	string	Model ID used to generate the response, like gpt-4o or o3. OpenAI offers a wide range of models with different capabilities, performance characteristics, and price points.
input	string	Text, image, or file inputs to the model, used to generate a response. Can be a string for simple text input, or an array of input items for multi-turn and multimodal inputs.
instructions	string	A system (or developer) message inserted at the beginning of the model's context. Use this for top-level guidance on the model's behavior, tone, or constraints.
previous_response_id	string	The unique ID of the previous response to the model. Use this to create multi-turn conversations. The model will use the previous response as context.
max_output_tokens	integer	An upper bound for the number of tokens that can be generated for a response, including visible output tokens and reasoning tokens.
temperature	number	What sampling temperature to use, between 0 and 2. Higher values like 0.8 make the output more random, while lower values like 0.2 make it more focused and deterministic.
top_p	number	An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass.
tools	array	An array of tools the model may use to generate a response. Supports built-in tools (web_search_preview, file_search, code_interpreter, computer_use_preview) and custom function tools.
tool_choice	string	How the model should select which tool to use. auto allows the model to decide, required forces the model to use a tool, none disables tool use.
truncation	string	The truncation strategy to use for the model context. auto will use the model-defined default truncation strategy. disabled will error if the context exceeds the model's context window.
text	object	Configuration for text response format.
reasoning	object	Configuration for reasoning models (o-series). Controls how much reasoning effort the model uses.
store	boolean	Whether to store the generated response for later retrieval via the GET /responses endpoint. Defaults to true.
metadata	object	Set of 16 key-value pairs that can be attached to an object. Useful for storing additional information in a structured format.
stream	boolean	If set to true, the model response data will be streamed as server-sent events to the client.
parallel_tool_calls	boolean	Whether to allow the model to run tool calls in parallel.
user	string	A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse.

View JSON Schema on GitHub

JSON Schema

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "title": "CreateResponseRequest",
  "type": "object",
  "properties": {
    "model": {
      "type": "string",
      "description": "Model ID used to generate the response, like gpt-4o or o3.\nOpenAI offers a wide range of models with different capabilities,\nperformance characteristics, and price points.\n"
    },
    "input": {
      "type": "string",
      "description": "Text, image, or file inputs to the model, used to generate\na response. Can be a string for simple text input, or an\narray of input items for multi-turn and multimodal inputs.\n"
    },
    "instructions": {
      "type": "string",
      "description": "A system (or developer) message inserted at the beginning\nof the model's context. Use this for top-level guidance\non the model's behavior, tone, or constraints.\n"
    },
    "previous_response_id": {
      "type": "string",
      "description": "The unique ID of the previous response to the model. Use\nthis to create multi-turn conversations. The model will\nuse the previous response as context.\n"
    },
    "max_output_tokens": {
      "type": "integer",
      "description": "An upper bound for the number of tokens that can be generated\nfor a response, including visible output tokens and reasoning\ntokens.\n"
    },
    "temperature": {
      "type": "number",
      "description": "What sampling temperature to use, between 0 and 2. Higher\nvalues like 0.8 make the output more random, while lower\nvalues like 0.2 make it more focused and deterministic.\n"
    },
    "top_p": {
      "type": "number",
      "description": "An alternative to sampling with temperature, called nucleus\nsampling, where the model considers the results of the\ntokens with top_p probability mass.\n"
    },
    "tools": {
      "type": "array",
      "description": "An array of tools the model may use to generate a response.\nSupports built-in tools (web_search_preview, file_search,\ncode_interpreter, computer_use_preview) and custom function\ntools.\n"
    },
    "tool_choice": {
      "type": "string",
      "description": "How the model should select which tool to use. auto allows\nthe model to decide, required forces the model to use a\ntool, none disables tool use.\n"
    },
    "truncation": {
      "type": "string",
      "description": "The truncation strategy to use for the model context. auto\nwill use the model-defined default truncation strategy.\ndisabled will error if the context exceeds the model's\ncontext window.\n"
    },
    "text": {
      "type": "object",
      "description": "Configuration for text response format."
    },
    "reasoning": {
      "type": "object",
      "description": "Configuration for reasoning models (o-series). Controls\nhow much reasoning effort the model uses.\n"
    },
    "store": {
      "type": "boolean",
      "description": "Whether to store the generated response for later retrieval\nvia the GET /responses endpoint. Defaults to true.\n"
    },
    "metadata": {
      "type": "object",
      "description": "Set of 16 key-value pairs that can be attached to an object.\nUseful for storing additional information in a structured format.\n"
    },
    "stream": {
      "type": "boolean",
      "description": "If set to true, the model response data will be streamed\nas server-sent events to the client.\n"
    },
    "parallel_tool_calls": {
      "type": "boolean",
      "description": "Whether to allow the model to run tool calls in parallel."
    },
    "user": {
      "type": "string",
      "description": "A unique identifier representing your end-user, which can\nhelp OpenAI to monitor and detect abuse.\n"
    }
  }
}