Responses

Create response (OpenAI Responses format)

Create a response in OpenAI Responses format. Input can be text or typed items; tools use the flat Responses format. Only POST /v1/responses is exposed: the Gateway does not expose response retrieval, deletion or conversation-management routes. Native dispatch preserves provider-specific JSON fields when the destination supports the caller’s format. Cross-format routing converts through the Gateway’s shared representation: provider-only features are not guaranteed to survive. Support for reasoning, images, built-in tools and structured outputs depends on the destination model.

post/v1/responses

Request

Base URL: https://edgee.io
URL: https://edgee.io/v1/responses
Auth: HTTP bearer

Headers

X-Edgee-Tagsstring

Comma-separated analytics tags. Trimmed and merged with body tags.

X-Edgee-Debug'true' | '1'

Enable debug capture for this request. Does not add a debug object to the API response.

X-Edgee-Compression-Tool-Result-Trimming'true' | 'false' | '1' | '0' | 'on' | 'off'

Turn tool-result trimming on or off. Overrides the API key setting for this request only. Values are case-insensitive, and a missing or invalid value keeps the key's setting.

X-Edgee-Compression-Tool-Surface-Reduction'true' | 'false' | '1' | '0' | 'on' | 'off'

Turn MCP tool surface reduction on or off. The threshold and resolver mode still come from the API key. Overrides the API key setting for this request only. Values are case-insensitive, and a missing or invalid value keeps the key's setting.

X-Edgee-Compression-Brevity'true' | 'false' | '1' | '0' | 'on' | 'off'

Turn output brevity on or off. Overrides the API key setting for this request only. Values are case-insensitive, and a missing or invalid value keeps the key's setting.

X-Edgee-Session-Idstring

Session identifier used to group request usage.

Request body

modelstring required

Model identifier from GET /v1/models, in author/model format, or a configured global alias. Append :provider to request a specific provider route. Available models and providers depend on your key and organization policies.

instructionsstring

System-level instruction prepended to the conversation. An alternative to including a system role message in the input array.

streamboolean

If set, the response is streamed as Server-Sent Events (SSE).

max_output_tokensinteger

Maximum number of tokens to generate.

temperaturenumber

Sampling temperature. Range and support depend on the destination model.

top_pnumber

Nucleus sampling probability. Alternative to temperature.

tagsstring[]

Optional tags to categorize and label the request. Useful for filtering and grouping requests in analytics and logs. Can also be sent via the x-edgee-tags header as a comma-separated string.

enable_debugboolean

Enable request debug capture. Gateway-only field; does not add debug information to the response.

Example request

{
  "model": "openai/gpt-5.2",
  "input": [
    {
      "call_id": "call_abc123",
      "name": "get_weather",
      "arguments": "{\"location\": \"Paris\"}"
    }
  ],
  "reasoning": {
    "effort": "high"
  }
}

Response

Response created successfully

idstring required

Unique identifier for the response, prefixed with resp_.

object'response' required

statusstring required

Shared converted responses use completed or in_progress; native responses can also report provider statuses such as incomplete or failed.

created_atnumber required

Unix timestamp (as a float) of when the response was created.

modelstring required

The model used to generate the response.

Example response

{
  "id": "resp_abc123",
  "created_at": 1677652288,
  "model": "openai/gpt-5.2"
}

Changes

No recorded changes to this operation across all 1 revision of this API.