GeekAI API does not limit user concurrency, and we strive to ensure service quality for all your requests.

Please note, however, that when our servers are under high traffic pressure, after your request is sent, it may take some time before receiving a server response. During this period, your HTTP request will maintain the connection and continuously receive content in the following format:

  • Non-streaming requests: continuous empty lines
  • Streaming requests: continuous SSE keep-alive comments (: keep-alive)

These contents do not affect the OpenAI SDK’s parsing of the JSON body response. If you are parsing HTTP responses yourself, please be mindful when handling these empty lines or comments.

If after 5 minutes the request is still not completed, the server will close the connection.