Rate limit error when not over the rate limit

I am routinely geting 429 rate limit responses, but the response from OpenAI itself doesn’t show that I am over the rate limit. For example:

Rate limit reached for default-gpt-3.5-turbo in organization org-xxxx on tokens per min. Limit: 90000 / min. Current: 86439 / min.

In fact, I have never seen one of these where the current rate limit is at or above the actual limit. Does anyone know what is going on with that?

This is happening to me as well. I’m well below the rate limit, but I still keep seeing the 429 errors. Did you figure this out?

If you look at the response headers, can you share the headers that start with x-ratelimit