API reference
Use inference endpoints for model requests.
Inference endpoints
| Method | Path | Purpose |
|---|---|---|
GET | /models | List accessible models. |
POST | /chat/completions | Send an OpenAI Chat Completions request. |
POST | /v1/responses | Send an OpenAI Responses request. |
POST | /v1/messages | Send an Anthropic Messages request. |
Send the API key as a bearer token:
Management API
Use the management API to manage workspaces, billing limits, workspace members and model availability.