Ratelimits on APIs

As per the official documentation, the current rate limits are:

* Quote API: 1 request/second
* Historical Candle API: 3 requests/second

What’s not clear is whether these limits are purely per-second throttles or if there’s an additional cap over a longer duration.

Do we have any upper bound like a rolling window limit (for example, 2000 requests per 30 minutes or per hour)?

We need clarity on this before designing anything that depends on sustained high-frequency calls.

Thanks