Skip to content

Rate limiting

Most API endpoints are subject to rate limiting. With rate-limiting, clients can only perform a certain number of requests within a specific time limit.

When rate-limiting is active on an endpoint, every response will include a set of rate-limiting headers that allow the client to know the current limit status.

Rate-limiting headers

  • X-Rate-Limit-Limit: length of the limited period.
  • X-Rate-Limit-Remaining: number of requests allowed within the period.
  • X-Rate-Limit-Reset: timestamp of the next upcoming limit reset. The client can leverage this value to know when it can perform an unlimited request.

When the rate limit has been reached, subsequent requests within the period obtain a 429 Too Many Requests response. Clients will have to wait for the next reset before performing successful requests.

Note

The limit is currently set to 100 requests per second across all the API surfaces. This value is subject to change at any moment. Therefore, clients should consider and leverage the rate-limiting information in every response.

Rate-limiting example

In the example below, the period is 1 second, with 99 requests left. The period will reset at 12:36:55 GMT, August 30, 2022:

X-Rate-Limit-Limit: 1s
X-Rate-Limit-Remaining: 99
x-Rate-Limit-Reset: 2022-08-30T12:36:55.2612963Z