GET

/v1/schemas/fal/chat/openrouter%2Frouter%2Fopenai%2Fv1%2Fresponses

Accept:

kind inputactivity chathash 9fa1cffe…extracted Jul 21, 2026version currentsource upstream spec ↗

Request body

0 properties · 0 required · 0 $defs

This schema isn’t a plain object with properties — showing the raw JSON instead.

application/json

{
  "provider": "fal",
  "activity": "chat",
  "endpointId": "openrouter/router/openai/v1/responses",
  "kind": "input",
  "contentHash": "9fa1cffea750963aaa272f233ea04b9f31fad190ace36850fda683ef304bd836",
  "specRevision": null,
  "provenance": {
    "sourceUrl": "https://api.fal.ai/v1/models?expand=openapi-3.0#openrouter/router/openai/v1/responses",
    "sourceHash": "e16a14d119efea96e45a99a20396613cd019ee5c81e8cfd618a4c5b619f07b5b",
    "fetchedAt": 1784610160,
    "extractorVersion": "1"
  },
  "createdAt": 1784610160,
  "supersededAt": null,
  "schema": {
    "additionalProperties": true,
    "description": "OpenAI-format request. This endpoint is an OpenAI-compatible passthrough to OpenRouter's Responses API: the request body follows the OpenAI Responses wire format and is forwarded with minimal normalization (usage accounting is enabled, and when a fal caller id is available, a privacy-preserving hash of it replaces any caller-supplied `user`); the response follows the OpenAI wire format except that `usage`, when returned non-null, is replaced with fal's accounting shape (`prompt_tokens`, `completion_tokens`, `total_tokens`, `cost`, `prompt_tokens_details`; fields may be null). When `stream` is true on a synchronous invocation, the response is delivered as server-sent events instead of a single JSON object. Authoritative behavior, including which fields each model accepts, is defined by OpenRouter's implementation (https://openrouter.ai/docs/api-reference); see also the OpenAI Responses API reference (https://platform.openai.com/docs/api-reference/responses). Any field accepted there is accepted here, subject to the normalization above.",
    "type": "object"
  }
}

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fal","endpointId":"openrouter/router/openai/v1/responses","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}