Skip to main content
POST
/
v2
/
gateway
/
responses
Create response
curl --request POST \
  --url https://api.orq.ai/v2/gateway/responses \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "input": "<string>",
  "metadata": {},
  "temperature": 1,
  "top_p": 0.5,
  "previous_response_id": "<string>",
  "instructions": "<string>",
  "reasoning": {
    "effort": "low"
  },
  "max_output_tokens": 123,
  "text": {
    "format": {
      "type": "text"
    }
  },
  "include": [
    "code_interpreter_call.outputs"
  ],
  "parallel_tool_calls": true,
  "store": true,
  "tools": [
    {
      "type": "function",
      "name": "<string>",
      "parameters": {
        "type": "object",
        "properties": {},
        "required": [
          "<string>"
        ],
        "additionalProperties": true
      },
      "description": "<string>",
      "strict": true
    }
  ],
  "tool_choice": "none",
  "stream": false
}
'
{
  "id": "<string>",
  "object": "response",
  "created_at": 123,
  "status": "completed",
  "error": {
    "code": "<string>",
    "message": "<string>"
  },
  "incomplete_details": {
    "reason": "max_output_tokens"
  },
  "model": "<string>",
  "output": [
    {
      "id": "<string>",
      "type": "message",
      "role": "assistant",
      "status": "in_progress",
      "content": []
    }
  ],
  "parallel_tool_calls": true,
  "instructions": "<string>",
  "output_text": "<string>",
  "usage": {
    "input_tokens": 123,
    "output_tokens": 123,
    "total_tokens": 123,
    "input_tokens_details": {
      "cached_tokens": 123
    },
    "output_tokens_details": {
      "reasoning_tokens": 123,
      "accepted_prediction_tokens": 123,
      "rejected_prediction_tokens": 123
    }
  },
  "temperature": 123,
  "top_p": 123,
  "max_output_tokens": 123,
  "previous_response_id": "<string>",
  "metadata": {},
  "tool_choice": "none",
  "tools": [
    {
      "type": "function",
      "name": "<string>",
      "parameters": {
        "type": "object",
        "properties": {},
        "required": [
          "<string>"
        ],
        "additionalProperties": true
      },
      "description": "<string>",
      "strict": true
    }
  ],
  "reasoning": {
    "effort": "<string>",
    "summary": "<string>"
  },
  "store": true,
  "text": {
    "format": {
      "type": "text"
    }
  },
  "truncation": "disabled",
  "user": "<string>",
  "service_tier": "auto",
  "background": true,
  "top_logprobs": 10,
  "logprobs": true
}

Authorizations

​
Authorization
string
header
required

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

Body

application/json
​
model
string
required

ID of the model to use. You can use the List models API to see all of your available models.

​
input
required

The actual user input(s) for the model. Can be a simple string, or an array of structured input items (messages, tool outputs) representing a conversation history or complex input.

​
metadata
object

Developer-defined key-value pairs that will be included in response objects

​
temperature
number | null

What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

Required range: 0 <= x <= 2
​
top_p
number | null

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

Required range: 0 <= x <= 1
​
previous_response_id
string | null

The ID of a previous response to continue the conversation from. The model will have access to the previous response context.

​
instructions
string | null

Developer-provided instructions that the model should follow. Overwrites the default system message.

​
reasoning
object

Configuration for reasoning models

​
max_output_tokens
integer | null

The maximum number of tokens that can be generated in the response

​
text
object
​
include
enum<string>[] | null

Specifies which (potentially large) fields to include in the response. By default, the results of Code Interpreter and file searches are excluded. Available options:

  • code_interpreter_call.outputs: Include the outputs of Code Interpreter tool calls
  • computer_call_output.output.image_url: Include the image URLs from computer use tool calls
  • file_search_call.results: Include the results of file search tool calls
  • message.input_image.image_url: Include URLs of input images
  • message.output_text.logprobs: Include log probabilities for output text (when logprobs is enabled)
  • reasoning.encrypted_content: Include encrypted reasoning content for reasoning models
Available options:
code_interpreter_call.outputs,
computer_call_output.output.image_url,
file_search_call.results,
message.input_image.image_url,
message.output_text.logprobs,
reasoning.encrypted_content
​
parallel_tool_calls
boolean | null

Whether to enable parallel function calling during tool use.

​
store
boolean | null
default:true

Whether to store this response for use in distillations or evals.

​
tools
object[]

A list of tools the model may call. Use this to provide a list of functions the model may generate JSON inputs for.

A function tool definition

  • Option 1
  • Option 2
  • Option 3
​
tool_choice

How the model should select which tool (or tools) to use when generating a response. Can be a string (none, auto, required) or an object to force a specific tool.

Available options:
none,
auto,
required
​
stream
boolean
default:false

Response

Returns a response object or a stream of events.

Represents the completed model response returned when stream is false

​
id
string
required

The unique identifier for the response

​
object
enum<string>
required

The object type, which is always "response"

Available options:
response
​
created_at
number
required

The Unix timestamp (in seconds) of when the response was created

​
status
enum<string>
required

The status of the response

Available options:
completed,
failed,
in_progress,
incomplete
​
error
object
required

The error that occurred, if any

​
incomplete_details
object
required

Details about why the response is incomplete

​
model
string
required

The model used to generate the response

​
output
object[]
required

The list of output items generated by the model

An assistant message output

  • Option 1
  • Option 2
  • Option 3
  • Option 4
​
parallel_tool_calls
boolean
required
​
instructions
string | null

The instructions provided for the response

​
output_text
string | null

A convenience field with the concatenated text from all text content parts

​
usage
object

Usage statistics for the response

​
temperature
number | null
​
top_p
number | null
​
max_output_tokens
integer | null
​
previous_response_id
string | null
​
metadata
object
​
tool_choice

Controls which (if any) tool is called by the model

Available options:
none,
auto,
required
​
tools
object[]

A function tool definition

  • Option 1
  • Option 2
  • Option 3
​
reasoning
object
​
store
boolean
​
text
object
​
truncation
enum<string> | null
default:disabled

Controls how the model handles inputs longer than the maximum token length

Available options:
auto,
disabled
​
user
string | null

A unique identifier representing your end-user

​
service_tier
enum<string> | null

The service tier used for processing the request

Available options:
auto,
default
​
background
boolean | null

Whether the response was processed in the background

​
top_logprobs
integer | null

The number of top log probabilities to return for each output token

Required range: 0 <= x <= 20
​
logprobs
boolean | null

Whether to return log probabilities of the output tokens