Home / Topics / Predictions
Rate limits
Table of contents
We limit the number of API requests that can be made to Replicate:
- You can create predictions at 600 requests per minute.
- All other endpoints you can call at 3000 requests per minute.
Throttling
You can make short bursts of requests above the default rate limits before being throttled.
As you approach running out of credit, we apply stronger rate limits. We do this to stop you from accidentally overspending and going into arrears, and to give you some time to increase your balance when it’s running low before getting shut off entirely.
To avoid this, set up credit auto-reload to keep your credit balance above $20.
API response
If you hit a limit, you will receive a response with status 429
with a body like:
{"detail":"Request was throttled. Your rate limit resets in ~30s."}
Higher limits
If you want higher limits, contact us.