Responses Response

The response object

Represents a response returned by the model in the Responses API, based on the provided input.

id `string`

Unique response identifier (e.g., resp-abc123def456).

object `string`

The object type, always "response".

created_at `integer`

Unix timestamp (in seconds) of the response creation.

status `string`

Response status: "completed", "failed", or "in_progress".

model `string`

The model used to generate the response.

output `array`

List of output items. Each item can be a reasoning item, a message or a function call. On S5 models the reasoning item comes first; to read only the answer text, use output_text in the OpenAI SDK.

reasoning (object)

The model's reasoning (S5 models). The text is confidential and only comes encrypted; on S5-nano (open weights), it comes as plain text in content, with no encrypted_content. Pass the item back on the next call so the model can resume its reasoning. See Reasoning.

id (string): Unique item identifier (rs_...).
type (string): Always "reasoning".
content (array): Empty ([]), except on S5-nano, where it carries the reasoning text:
- type (string): "reasoning_text".
- text (string): The reasoning text.
encrypted_content (string): Encrypted reasoning. Can only be used by the same account. Absent on S5-nano.
summary (array): Reasoning summary, when requested in reasoning.summary; otherwise empty.
- type (string): "summary_text".
- text (string): Summary text.

message (object)

id (string): Unique message identifier.
type (string): Always "message".
role (string): Always "assistant".
status (string): "completed" or "in_progress".
content (array): List of content parts:
- type (string): "output_text".
- text (string): Text generated by the model.
- annotations (array): Annotations (usually empty).

function_call (object)

id (string): Unique item identifier.
type (string): Always "function_call".
name (string): Name of the function to call.
arguments (string): Arguments as serialized JSON.
call_id (string): Call identifier (used to send the result back).
status (string): "completed".

reasoning `object or null`

Reasoning configuration used (S5 models). Does not contain the reasoning text.

Show properties

effort (string): Effort used (xhigh when not set).
summary (string or null): Requested summary type (auto shows up as detailed).
context (string): all_turns or current_turn.
mode (string): Always "standard".

previous_response_id `string or null`

Id of the previous response used in the request, if any.

usage `object`

Token usage statistics.

Show properties

input_tokens (integer): Tokens consumed in the input.
output_tokens (integer): Tokens generated in the output.
total_tokens (integer): Total tokens (input + output).
input_tokens_details (object or null):
- cached_tokens (integer): Tokens from cache.
output_tokens_details (object or null):
- reasoning_tokens (integer): Tokens used for reasoning.

error `object or null`

Error object when status is "failed".

Show properties

code (string): Error code (e.g., "server_error", "internal_error").
message (string): Descriptive error message.

incomplete_details `object or null`

Details about why the response is incomplete (e.g., reached max_output_tokens).

Show properties

reason (string): Interruption reason (e.g., "max_output_tokens").

instructions `string or null`

System instructions sent in the request.

temperature `float or null`

Sampling temperature used.

top_p `float or null`

Top-p used.

max_output_tokens `int or null`

Configured token limit.

metadata `object`

Request metadata (key-value pairs).

tools `array`

Tools that were available to the model.

tool_choice `string or object or null`

Tool usage configuration.

parallel_tool_calls `boolean`

Whether parallel tool calls were enabled.

store `boolean`

Whether the response was stored (default true). Stored responses are available for 30 days for previous_response_id.

Streaming Events

When stream: true, the API returns events via Server-Sent Events (SSE).

Event	Description
`response.created`	Response created
`response.in_progress`	Generation started
`response.output_item.added`	New output item added
`response.content_part.added`	New content part started
`response.output_text.delta`	Text fragment generated
`response.output_text.done`	Full text finalized
`response.content_part.done`	Content part finalized
`response.output_item.done`	Output item finalized
`response.reasoning_summary_part.added`	New reasoning summary part started (with `reasoning.summary`)
`response.reasoning_summary_text.delta`	Reasoning summary fragment
`response.reasoning_summary_text.done`	Reasoning summary finalized
`response.reasoning_summary_part.done`	Reasoning summary part finalized
`response.completed`	Response complete (with `usage`)
`response.failed`	Error during generation

All events include a sequence_number field incremented with each event.

On S5 models, the reasoning item is the first output item (output_index 0) and already carries encrypted_content in response.output_item.added and response.output_item.done. There are no events with the reasoning text. On S5-nano there is no encrypted_content: the full reasoning text comes in the reasoning item of response.output_item.done (in added, content is empty).

HTTP Error Codes

Status	Description
`200`	Success
`400`	Unsupported parameter or invalid request
`401`	Missing or invalid API key
`422`	Validation error
`429`	Rate limit exceeded
`500`	Internal server error

Response Example

{
  "id": "resp-abc123def456",
  "object": "response",
  "created_at": 1234567890,
  "status": "completed",
  "model": "sabia-4",
  "output": [
    {
      "id": "msg-def456",
      "type": "message",
      "role": "assistant",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "text": "The capital of Brazil is Brasília.",
          "annotations": []
        }
      ]
    }
  ],
  "usage": {
    "input_tokens": 12,
    "output_tokens": 8,
    "total_tokens": 20,
    "input_tokens_details": {
      "cached_tokens": 0
    },
    "output_tokens_details": {
      "reasoning_tokens": 0
    }
  },
  "temperature": 0.7,
  "top_p": 0.95,
  "max_output_tokens": null,
  "instructions": null,
  "metadata": {},
  "tools": [],
  "tool_choice": "auto",
  "parallel_tool_calls": true,
  "store": true,
  "error": null,
  "incomplete_details": null
}

Example with reasoning (S5)

{
  "id": "resp-7c1e9a2b",
  "object": "response",
  "status": "completed",
  "model": "S5",
  "output": [
    {
      "id": "rs_6f1c0d3e",
      "type": "reasoning",
      "content": [],
      "encrypted_content": "gAAAAABp...",
      "summary": []
    },
    {
      "id": "fc-abc123",
      "type": "function_call",
      "name": "get_weather",
      "arguments": "{\"city\": \"Campinas\"}",
      "call_id": "call_abc123",
      "status": "completed"
    }
  ],
  "reasoning": {
    "effort": "xhigh",
    "summary": null,
    "context": "all_turns",
    "mode": "standard"
  },
  "previous_response_id": null,
  "store": true,
  "usage": {
    "input_tokens": 52,
    "output_tokens": 230,
    "total_tokens": 282,
    "output_tokens_details": {
      "reasoning_tokens": 204
    }
  }
}

function_call output example

{
  "id": "resp-xyz789",
  "object": "response",
  "created_at": 1234567890,
  "status": "completed",
  "model": "sabia-4",
  "output": [
    {
      "id": "fc-abc123",
      "type": "function_call",
      "name": "get_weather",
      "arguments": "{\"city\": \"São Paulo\"}",
      "call_id": "call_abc123",
      "status": "completed"
    }
  ],
  "usage": {
    "input_tokens": 25,
    "output_tokens": 15,
    "total_tokens": 40
  }
}

Streaming Example (SSE)

event: response.created
data: {"type":"response.created","response":{...},"sequence_number":1}

event: response.in_progress
data: {"type":"response.in_progress","response":{...},"sequence_number":2}

event: response.output_item.added
data: {"type":"response.output_item.added","output_index":0,"item":{...},"sequence_number":3}

event: response.content_part.added
data: {"type":"response.content_part.added","item_id":"msg-def456","output_index":0,"content_index":0,"part":{"type":"output_text","text":""},"sequence_number":4}

event: response.output_text.delta
data: {"type":"response.output_text.delta","item_id":"msg-def456","output_index":0,"content_index":0,"delta":"The capital","sequence_number":5}

event: response.output_text.delta
data: {"type":"response.output_text.delta","item_id":"msg-def456","output_index":0,"content_index":0,"delta":" of Brazil","sequence_number":6}

event: response.output_text.done
data: {"type":"response.output_text.done","item_id":"msg-def456","output_index":0,"content_index":0,"text":"The capital of Brazil is Brasília.","sequence_number":7}

event: response.content_part.done
data: {"type":"response.content_part.done","item_id":"msg-def456","output_index":0,"content_index":0,"part":{"type":"output_text","text":"The capital of Brazil is Brasília."},"sequence_number":8}

event: response.output_item.done
data: {"type":"response.output_item.done","output_index":0,"item":{...},"sequence_number":9}

event: response.completed
data: {"type":"response.completed","response":{...,"usage":{"input_tokens":12,"output_tokens":8,"total_tokens":20}},"sequence_number":10}

The response object

id string​

object string​

created_at integer​

status string​

model string​

output array​

reasoning object or null​

previous_response_id string or null​

usage object​

error object or null​

incomplete_details object or null​

instructions string or null​

temperature float or null​

top_p float or null​

max_output_tokens int or null​

metadata object​

tools array​

tool_choice string or object or null​

parallel_tool_calls boolean​

store boolean​

Streaming Events​

HTTP Error Codes​

Response Example​

Example with reasoning (S5)​

function_call output example​

Streaming Example (SSE)​

id `string`

object `string`

created_at `integer`

status `string`

model `string`

output `array`

reasoning `object or null`

previous_response_id `string or null`

usage `object`

error `object or null`

incomplete_details `object or null`

instructions `string or null`

temperature `float or null`

top_p `float or null`

max_output_tokens `int or null`

metadata `object`

tools `array`

tool_choice `string or object or null`

parallel_tool_calls `boolean`

store `boolean`

Streaming Events

HTTP Error Codes

Response Example

Example with reasoning (S5)

function_call output example

Streaming Example (SSE)