Skip to main content
OpenPAI exposes three protocol endpoints at once, so you can choose freely based on your existing code or the upstream model.

​
Endpoint quick reference

The same sk- key works across all three endpoints, sharing the account quota.

​
API categories

OpenAI-compatible

/v1/chat/completions, /v1/embeddings, /v1/images/*, /v1/audio/*, /v1/responses, /v1/realtime

Claude Messages

/v1/messages, /v1/messages/count_tokens

Gemini native

:generateContent, :streamGenerateContent, :embedContent

Rerank

/v1/rerank — Cohere / Jina compatible

Midjourney

/mj/submit/*, /mj/task/{id}/fetch

Suno

/suno/submit/music, /suno/fetch/{id}

​
Common conventions

  • Protocol: HTTPS, HTTP/2 recommended (curl --http2).
  • Encoding: Content-Type: application/json; charset=utf-8.
  • Timeout: server default 600 seconds. Streaming responses keep a long connection, with an idle heartbeat every 15 seconds.
  • Request body size: default limit 32 MB (after compression).
  • Response headers:
    • X-OpenPAI-Request-Id: the request ID — include it in support tickets to speed up diagnosis.
    • X-OpenPAI-Group: the user group this request used.
    • X-OpenPAI-Quota-Consumed: the amount charged for this request.
    • X-OpenPAI-Remaining-Quota: the account balance after the charge.

​
Next steps

Authentication

Auth header formats and examples for the three protocols.

Error codes

HTTP statuses, error structure and troubleshooting tips.

Rate limits

RPM / TPM limits and handling 429.