Common API

API version: v1.2.4

Public contract used by all external Phosor AI APIs.

Base URLs:

All requests require X-API-Key unless the endpoint is explicitly public.

Contract Rules

request_id is the only public async identifier.
Async submit endpoints return accepted state first, then callers poll by request_id.
Public responses prefer flat fields over nested wrapper objects.
Error responses use a short machine-readable error plus a human-readable message.

JSON

{
  "error": "string",
  "message": "string",
  "detail": "optional string"
}

Accepted response:

JSON

{ "request_id": "uuid", "status": "pending" }

Queued / processing task detail:

JSON

{
  "request_id": "uuid",
  "status": "pending",
  "result": {},
  "error": null
}

Each account may have a limited number of Model API inference jobs running at once. Exceeding it returns 429 — wait for a job to finish and retry.

The limit applies to Model API inference (/api/v1/inference/submit). Image Studio runs on its own pool and does not consume this quota.