Requests per minute
Each key can make 600 requests per minute.whoami reports the limit in effect as rate_limit.requests_per_minute.
Every response carries the key’s budget:
| Header | Meaning |
|---|---|
X-RateLimit-Limit | Requests allowed per minute |
X-RateLimit-Remaining | Requests left in the current window |
X-RateLimit-Reset | Unix time, in seconds, by which the full budget is available again |
Concurrent requests
At most 4 requests per key, and 8 per organization across all its keys, can be in progress at once. A request beyond that returns429 with Retry-After: 1. Page through results one request at a time per product.
Failed authentication
60 failed authentications from one IP address in 10 minutes block further requests from that address until the window ends. Successful requests don’t count, so many integrations sharing one outbound address are unaffected while their keys are valid.When you are limited
Limited requests return429 with code: rate_limited and a Retry-After header, in seconds. Wait that long before retrying.
A single request that runs too long fails rather than being limited. If a large page fails with a 5xx, retry with a smaller limit or a narrower filter.