Run inference, tokenize, or create a file/batch
Covers v1/chat/completions, v1/responses, v1/responses/compact (compacts a prior
Responses API input window for a subsequent turn), and v1/tokenize-text, plus creating a file or batch
(registered to your account on success — a batch’s own per-item bodies face the identical
field policy a live call would). The request model must be on the gateway’s catalog,
user is always overwritten with a per-tenant identifier, and tools[] is checked by its
explicit type — see the Grok tag description above.
Authentication
Your per-client bearer token — the API key the gateway issued to you. Present it in
Authorization: Bearer <token> form. A token that is missing, unknown, or revoked draws a
401. Before the call is sent upstream the token is removed (stripped, or swapped for the
upstream’s own credential); it never makes it past the gateway.
Path parameters
The path to call on xAI’s Grok API, any further / segments included — all of it after
/grok/ is passed to https://api.x.ai/ when the gateway’s model catalog and
endpoint/field policy admit the request (see the Grok tag description above). Examples:
v1/chat/completions, v1/chat/deferred-completion/{id}, v1/responses,
v1/responses/compact, v1/tokenize-text, v1/models, v1/models/{model}, v1/files,
v1/files/{id}, v1/batches, v1/batches/{id}.
Request
Response
Whatever the upstream answered, relayed as-is (status, headers, body). Rendered here as JSON; the true shape is the upstream provider’s to define.