RAG API TPM

[rwmehta]

I saw that there was a feature request for this in 2023 Discussion 434, and I was wondering if anything came of it. I'm consistently running into token per minute limits when trying to upload even 500 KB CSVs and my embedding model is already set to its max of 350K tokens per minute. It would be great if there's a way to add delay to the calls and avoid this issue.