01/Overview
API Reference
The endpoints the gateway exposes and authentication.
All endpoints require a Bearer token. Interactive docs are available on the running server at /api/docs (Swagger) and /api/redoc.
#OpenAI-compatible
| Endpoint | Description |
|---|---|
| POST /v1/chat/completions | Chat completion (streaming supported). |
| POST /v1/completions | Legacy completion. |
| POST /v1/embeddings | Generate embeddings. |
| POST /v1/responses | Responses API. |
| GET /v1/models | List of reachable models/groups. |
#Ollama-native
| Endpoint | Description |
|---|---|
| POST /api/chat · /api/generate | Ollama chat/generate. |
| POST /api/embeddings · /api/embed | Embeddings. |
| GET /api/tags · POST /api/show | Model list / detail. |
#IDE-specific
Compatible paths are provided for tools like Cursor and Codex (/cursor/chat/completions, /codex/*). These work together with the matching node types (cursor, antigravity, bedrock).