Skip to main contentRate limits help protect the platform and ensure fair usage. This page explains how they function with our Gateway and providers.
Our approach to rate limits
We don’t impose our own rate limits on users. Instead, we collaborate closely with AI providers to secure the highest possible limits for our users. Any rate limits you encounter will come directly from the providers you’re being routed to. We hope to be able to publish information about how often limits are hit soon, but in the meantime, please reach out if you have specific questions/concerns.
If you use BYOK, then the rate limits set by your provider account will apply over our limits.
Handling rate limits
Continue to make requests as usual, we will attempt to automatically fallback to ensure you still have successful responses for every request.
Monitoring usage
- The AI Stats dashboard shows per-endpoint charts for requests, latency, and errors.
- Export usage metrics via the Dashboard and soon, the API, for any self reporting.
