Request Rate Limits

The API is protected by a rate-limiting system to prevent abuse. When too many requests are made to the API by a client, the server will respond with an HTTP 429 status code. The Retry-After header in the response should be used by clients to know how long to wait before attempting any new request.

Newly registered applications are set up with a limit of 5 requests/sec on all HTTP endpoints as well as 3 concurrent websocket connections by type (preprocessed/computed).

The purpose of these restrictions is for application developers to make sure that any errors and limitations related to the rate-limit are properly handled during development, before going to production. Once the API integration is deemed ready for deployment, or testing on a larger scale, developers are invited to contact our API support team in order to configure more appropriate limits. The production limits are decided on a case-by-case basis, depending on use cases.