Register the model with the LLM gateway
Make the deployment’s model available for inference: registers it in the LLM gateway’s model registry under the deployment’s name. Deployment must be ready. Registration is deliberately decoupled from deployment so pricing, budgets, and rate limits can be configured before the model becomes callable.
Authorizations
Gateway auth: send Authorization: Bearer <mka1-api-key>. For multi-user server-side integrations, you can also send X-On-Behalf-Of: <external-user-id>.
Headers
Optional external end-user identifier forwarded by the API gateway.
Path Parameters
Response
Successful Response
A deployed inference server.
A hardware accelerator request — GPU, NPU, TPU, or similar.
Request an accelerator by type and count, with optional fallback types.
Inference engine that backs a deployment.
vllm, sglang Current config revision
Autoscaling configuration for a deployment.
pending, provisioning, ready, scaling, updating, degraded, failed, stopped, deleted Where and how to call the model; set while registered with the gateway
Model ID consumers pass to the gateway's inference endpoints; set while the deployment is registered with the gateway (see Register), null when unregistered