Rate limits
How many requests a key may make.
The limit
Each API key may make 60 requests a minute. The count is per key: two keys of the same workspace each have their own.
Headers
Every response to an accepted key says where it stands:
| Header | Meaning |
|---|---|
RateLimit-Limit | Requests the key may make in a window |
RateLimit-Remaining | Requests left in the current window |
RateLimit-Reset | Seconds until the window ends and the count starts again |
A response to a refused key (401) does not carry them.
Over the limit
Past the limit, the API answers 429 with the code rate_limit.exceeded and a Retry-After header: the seconds until the window ends. Wait that long, then send the request again.
Spread the requests out
A batch job that reads RateLimit-Remaining and pauses when it nears zero never sees a 429. One that sends everything at once spends the window in its first second and waits out the rest.