Rate limits

Limits are per credential, not per IP, so one noisy integration cannot starve another.

Limits and backoff

Limits are per credential, not per IP, so one noisy integration cannot starve another. The default ceiling is 120 requests per minute per token, and it can be lowered per token in the panel.

Over the limit you get 429. Back off exponentially and retry with the same Idempotency-Key — that is exactly the case the key exists for.