Rate limits¶
| Limit | Applies to | Scope |
|---|---|---|
| 600 requests a minute | Requests with a valid key | Per key |
| 60 failed authentications a minute | Requests with a missing or invalid key | Per IP address |
Going over either answers 429 rate_limited with a Retry-After header giving the seconds to wait.
The second limit refuses every request from that address until the minute passes, including requests with a valid key. A misconfigured client sharing an address with a working one can block both, so fix repeated 401s rather than retrying them.
Handle a 429¶
Wait for the time in Retry-After, then retry. The same applies to 503 quota_unavailable and cloud_unavailable, which also send it.
--retry covers 429 and 503, and curl waits for Retry-After when the server sends it.