Handle rate limit

Rate limit per endpoint

API allows 240 requests per minute from a single IP address. When the limit is exceeded, one will receive 429 HTTP responses. In such cases, the API will likely include a number of seconds to wait in the Retry-After response header. Please build an integration around the idea that API will throttle requests by default. This way, we can ensure that all users can reliably use the API.

Throttling in other services

API uses other services that impose their own rate limits. Therefore, it is expected that integration will implement the imposed rate limits defined in the Retry-After header accordingly.

When no Retry-After is given

In case when API is degraded or no clear indication of seconds to wait, it is best to retry a request in 60 seconds. If the issue persists, please report it.