From the beginning of the release of our API, we have had a rate limit that is set to 1000 requests/minute per accesstoken. If we compare with other API:s on the web, such as Github, Reddit or Twitch they have a rate limit towards their API that varies between 60-120 requests/minute.
Over the last couple of months we more and more often have experience very heavy load on our API and our database servers, which is related to the API and number of requests. Therefor we have decided to lower our rate limits to 200 request/minute.
We are doing this ensure the availability to all of our services.
At the same time, we will change the HTTP status code when the rate limit is hit, from 503 “Service Unavailable” to 429 “Too Many Requests”.
The change will take affect on the 4th of February 2019.