Rate Limit
The SUTRA API does not enforce rigid rate limits on usage. Our goal is to process all incoming requests efficiently.
That said, during periods of heavy server traffic, response times may be longer. While waiting, your HTTP request will stay open, and the server may provide the following updates to maintain the connection:
-
For non-streaming requests: Periodic empty lines will be sent.
-
For streaming requests: Server-Sent Events (SSE) comments like
: keep-alive
will be transmitted.
These interim updates are harmless and do not disrupt JSON parsing when using compatible SDKs. If you are handling responses manually, ensure your implementation can manage these empty lines or comments seamlessly.
In cases where a request remains incomplete for over 30 minutes, the server will close the connection automatically.