Rate Limit Advice - update

OpenAI FAQ - Rate Limit Advice - Update

Rate limits can be quantized, meaning they are enforced over shorter periods of time (e.g. 60,000 requests/minute may be enforced as 1,000 requests/second). Sending short bursts of requests or contexts (prompts+max_tokens) that are too long can lead to rate limit errors, even when you are technically below the rate limit per minute.

OpenAI FAQ - How can I solve 429: ‘Too Many Requests’ errors?

As unsuccessful requests contribute to your per-minute limit, continuously resending a request won’t work.

4 Likes

Good to know. Thanks for passing this along. So many things to keep track of these days, it can be easy to miss something like this!

1 Like

The thanks go to @elmstedt. I just had to scan them as they were created.

1 Like

Even that takes time!

"… 195 replies… " :wink:

Thanks to all, though.

1 Like

Things should, in theory, be fixed now so there won’t be trivial updates posted. I just hope Logan gets back and gives the go ahead for a new sub-category and an API key for it.

2 Likes