Rate Limits

This is a security/abuse-protection control, separate from entitlement quota. Rate limiting exists to keep the API healthy for everyone; quota is about what your Plan includes.

How it works

Requests are rate-limited per Credential. When the limit is exceeded, the API returns 429 RATE_LIMITED and may include a Retry-After header telling you how many seconds to wait.

Limiting is currently applied at the Credential level, shared across every /api/v1/* call that Credential makes — it is not currently broken out per individual endpoint. A burst of calls to one endpoint can affect the budget available to another call made with the same Credential shortly after.

What we don't publish

The exact numeric threshold is an internal operational setting, not a fixed public guarantee — it may be adjusted. Design your integration to handle 429/Retry-After gracefully (back off and retry) rather than assuming a specific number of requests per interval.

What to do on a 429

Wait for the duration indicated by Retry-After (if present), then retry. A 429 never consumes any of your entitlement quota — a rate-limited request costs you nothing on the quota side.