What happens when I reach my monthly request limit?
Updated
It depends on your plan. On Free, requests stop at 25,000 a month and you're never charged. On Pro and Scale, nothing stops; extra requests are billed at the end of your billing cycle.
Free plan#
- The 25,000 requests are a hard cap, with no overage charges.
- Once you reach it, the API returns a limit error until you upgrade or the next month starts.
- Upgrading takes effect right away, so your app recovers as soon as you do.
Pro and Scale#
- Your app keeps working past the included requests.
- Extra requests cost $10 per 100,000 on Pro and $6 per 100,000 on Scale.
- They're added to the invoice at the end of the billing cycle.
How do I avoid a surprise bill?#
- Watch usage on the console's Billing & usage page.
- Set spend limits per project or per workflow. They cap AI provider spend, and requests stop with a clear budget error when a cap is reached.
- Use test mode while developing; test requests don't count.
What's the difference between the monthly limit and the rate limit?#
The monthly limit is the total requests in a billing period. The rate limit is requests per minute, per project: 40 on Free, 200 on Pro and 1,000 on Scale. Going over the rate limit gives a 429 error for a moment, not a block for the month. Retry with backoff.