Audio, embeddings & non-chat endpoints
Auto-routing applies only to chat completions. All other endpoint types bypass the router entirely and reach their real models — even on an auto key. This matters because routing a Whisper request to a text model would fail.
| Endpoint | Behavior |
|---|---|
/v1/chat/completions | Auto-routed (on auto keys / router models) |
/v1/audio/transcriptions | Pass-through → whisper-1 / gpt-4o-transcribe (OpenAI) |
/v1/embeddings | Pass-through → text-embedding-3-small/-large, ada-002 (OpenAI) |
| Image generation, moderation, rerank, speech | Pass-through |
Practical effect: an application can use one auto key for everything — chat is cost-optimized by the router, while voice transcription and RAG embeddings just work.