Rate limit thresholds and how they are enforced
Rate limits vs compute units
Rate limits are based on requests per min and are fully independent of overall compute limit usage. Going over your compute units will not automatically trigger rate limits.
Rate limits are set per subscription plan
- Starter plan rate limits are 300 RPM or 5 RPS
- Growth plan rate limits are 600 RPM or 10 RPS
- Scale plan rate limits are 1200 RPM or 20 RPS
If you need higher limits, reach out and we can set up an enterprise plan for you.
API-specific rate limits in RPM
This type of rate limit is triggered when you call the same API above its API-specific threshold on your plan. As an example - On the Growth plan, you can call two different APIs at 500 RPM each for a total RPM of 1000 without triggering your rate limits.
| API | Starter (RPM) | Growth (RPM) | Scale (RPM) | Enterprise |
|---|---|---|---|---|
| POST v2/farcaster/frame/validate | 5k | 10k | 20k | Custom |
| GET v2/farcaster/signer | 3k | 6k | 12k | Custom |
| GET v2/farcaster/signer/developer_managed | 3k | 6k | 12k | Custom |
| All others | 300 | 600 | 1200 | Custom |
Global rate limits in RPM
This type of rate limit is triggered when you call any API above the global threshold on your plan.
/validate, /signer and signer/developer_managed APIs don't count towards global rate limits.
| API | Starter (RPM) | Growth (RPM) | Scale (RPM) | Enterprise |
|---|---|---|---|---|
| Across all APIs | 500 | 1000 | 2000 | Custom |