Run a model by id
Runs any model in the catalog by its public id, with input passed through untyped — the same call the typed operations below make, minus the compile-time schema.
Reach for it when the model is not known ahead of time: a client generated before a model shipped can still run it, and an id read from GET /v3/models at runtime needs no regeneration. Prefer the typed operation whenever your client already has one — input here is validated against the same published schema (GET /v3/models/{model}), so a bad field is a 400 at submit rather than an error before the call.
Submits an asynchronous job and returns 202 with a job id. Fetch the result at GET /v3/jobs/{job_id} — each item in its outputs[] follows the OutputItem schema — or track progress via GET /v3/jobs/{job_id}/status / the SSE stream at GET /v3/jobs/{job_id}/stream.
Authorizations
Primary scheme: Authorization: Key <key_id>:<secret>.
Path Parameters
The model's public id (GET /v3/models).
Body
Model-specific inputs, validated at submit against the model's published input schema (GET /v3/models/{model}).
URL to receive a signed completion webhook.
Replays the original ack for a retried submit instead of enqueueing a duplicate job.
Response
Accepted. The job runs asynchronously; poll status_url / result_url from the ack.
This job's id — server-issued, and opaque.
The resolved model id this job runs on.
The four states a job can be in.
IN_QUEUE, IN_PROGRESS, COMPLETED, FAILED Path of this job's status monitor: poll GET /v3/jobs/{job_id}/status for status, progress, and an estimate.
Path of the job resource itself: GET /v3/jobs/{job_id} returns the result envelope, including the outputs once it completes. Also the value of this response's Location header.
ISO-8601 instant this job is estimated to finish. Null when no estimate exists for the model yet; poll GET /v3/jobs/{job_id}/status for a refreshed one.