Skip to main content

​
Rate Limits

Limits are per API key and enforced with sliding windows.

​
Limit types

  • RPM — Requests Per Minute (any 60-second rolling window)
  • RPD — Requests Per Day (any 24-hour rolling window)
  • Some plans also track weekly request or token budgets.

​
What counts

A request counts toward limits when the final status recorded is:
  • success
  • client_error (4xx from client or bad request)
Upstream errors and capacity unavailable (5xx, 503 “no providers”) do not count.

​
Behavior on exceed

The API returns HTTP 429 Too Many Requests. The limit resets gradually as the sliding window moves forward.

​
Best practices

  • Use separate keys for bursty workloads.
  • Monitor usage in the dashboard.
  • Purchase credits or upgrade when you consistently hit limits.
  • For background jobs, add jitter and respect 429 Retry-After style backoff (when provided).
See your plan details on the Pricing page or in the dashboard.