Model Responses

Generate model responses using the specified model.

POST /responses

Model Responses

cURL

curl --request POST \
  --url https://{gatewayBaseURL}/responses \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "input": "<unknown>",
  "background": true,
  "include": [
    "<string>"
  ],
  "instructions": "<string>",
  "max_output_tokens": 123,
  "metadata": {},
  "parallel_tool_calls": true,
  "previous_response_id": "<string>",
  "reasoning": {
    "effort": "<string>"
  },
  "store": true,
  "stream": true,
  "temperature": 123,
  "text": {
    "format": {}
  },
  "tools": [
    "<unknown>"
  ],
  "top_p": 123,
  "user": "<string>"
}
'

200

{
  "id": "<string>",
  "object": "<string>",
  "created_at": 123,
  "status": "<string>",
  "max_output_tokens": 123,
  "model": "<string>",
  "output": [
    {
      "id": "<string>",
      "type": "<string>",
      "status": "<string>",
      "content": [
        {
          "type": "<string>",
          "annotations": [
            "<unknown>"
          ],
          "text": "<string>"
        }
      ],
      "role": "<string>"
    }
  ],
  "parallel_tool_calls": true,
  "previous_response_id": "<string>",
  "reasoning": {
    "effort": "<unknown>",
    "summary": "<unknown>"
  },
  "service_tier": "<string>",
  "store": true,
  "temperature": 123,
  "text": {
    "format": {
      "type": "<string>"
    }
  },
  "tool_choice": "<string>",
  "tools": [
    "<unknown>"
  ],
  "top_p": 123,
  "truncation": "<string>",
  "usage": {
    "input_tokens": 123,
    "input_tokens_details": {
      "cached_tokens": 123
    },
    "output_tokens": 123,
    "output_tokens_details": {
      "reasoning_tokens": 123
    },
    "total_tokens": 123
  },
  "metadata": {},
  "provider": "<string>",
  "error": "<unknown>",
  "incomplete_details": "<unknown>",
  "instructions": "<unknown>",
  "user": "<unknown>"
}

Authorizations

Authorization · string · header · required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Headers

x-tfy-metadata · string

Optional metadata for the request

Body

application/json

Parameters for generating model responses.

model · string · required

Model identifier to generate the response

input · any

Text, image, or file inputs to the model

background · boolean | null

Whether to run the model response in the background

include · string[] | null

Additional output data to include

instructions · string | null

System message as first item in context

max_output_tokens · number | null

Upper bound for tokens generated

metadata · object

Key-value pairs for additional information

Show child attributes

parallel_tool_calls · boolean | null

Allow parallel tool calls

previous_response_id · string | null

ID of previous response for multi-turn

reasoning · object

Configuration for reasoning models

Show child attributes

service_tier · enum<string> | null

Latency tier for processing

Available options: auto, default, flex, priority

store · boolean | null

Whether to store the response

stream · boolean | null

Enable streaming response

temperature · number | null

Sampling temperature between 0 and 2

text · object

Text response configuration

Show child attributes

tool_choice · Option 1 · enum<string> Option 2 · enum<string> Option 3 · enum<string> Option 4 · object Option 5 · any

Tool selection behavior

Available options: none

tools · any[] | null

Available tools for the model

top_p · number | null

Nucleus sampling parameter

truncation · enum<string> | null

Truncation strategy

Available options: auto, disabled

user · string | null

End-user identifier

Response

200

application/json

Model Response generated successfully.

id · string · required

Response ID.

object · string · required

Object type.

created_at · number · required

Creation timestamp.

status · string · required

Response status.

max_output_tokens · number | null · required

Maximum output tokens allowed.

model · string · required

Model used for the response.

output · object[] · required

Show child attributes

parallel_tool_calls · boolean · required

Indicates if parallel tool calls were used.

previous_response_id · string | null · required

ID of the previous response, if any.

reasoning · object · required

Reasoning details.

Show child attributes

service_tier · string · required

Service tier.

store · boolean · required

Indicates if the response is stored.

temperature · number · required

Temperature setting for the model.

text · object · required

Show child attributes

tool_choice · string · required

Tool choice used.

tools · any[] · required

Tools used in the response.

top_p · number · required

Top-p sampling parameter.

truncation · string · required

Truncation setting.

usage · object · required

Show child attributes

metadata · object · required

Additional metadata.

provider · string · required

Provider of the response.

error · any

Provider error object when the response failed.

incomplete_details · any

Reason the response was cut off, e.g. max_output_tokens or content_filter.

instructions · any

Instructions provided for the response.

user · any

End-user identifier echoed from the request.