Base URL
Authentication
Send your API key as a Bearer token on every request.Plans
Access is packaged into three plans. Every plan includes the current-data endpoints below. Paid plans add capacity, history, and delivery features.Plan enforcement, historical data, snapshots, incremental updates, webhooks,
and embeds are planned and roll out in stages. Today every key has access
to the six current-data endpoints under the existing route limits. Pages that
describe planned behavior are labeled, and the
changelog tracks what is live.
Endpoints at a glance
See the Endpoints section for full request and response schemas.
/v1/updates lists recently added models within a lookback window. It is
not a change feed: it does not report score, ranking, or pricing changes, and
it has no cursor. For durable creates, updates, and removals after a cursor,
see the incremental updates API.Discovering gateway-routable models
Every model returned by/v1/models and /v1/models/{id} carries an
inference block that tells you whether the model can be invoked
through the LLM Stats gateway and the
canonical id to pass as model:
available is false, the optional fields are omitted —
that model is in the catalog but cannot be called through the gateway.
Errors, limits, and headers
- Every error uses one envelope with a stable
error.code. See Errors. - Burst limits and daily quotas apply per organization, not per API key. Creating another key does not create another quota pool. See Rate limits and headers.