# Rate limits

All your API keys share one set of limits: by default 600 requests a minute, 32 runs in
progress and 32 open event streams (4 per user), plus daily and monthly AI spend. Read yours
with [Get your rate limits](/api-reference/partner/quotas).

When a limit is reached the API answers `429` with a `reason` and, when it can tell,
a `Retry-After` header. Wait and retry the same request. See
[Usage and limits](/usage-and-quotas) for the details.
