Incorrect error displaying when context size is larger than organization TPM

[asim-shrestha]

Describe the bug

When you make a request to the OpenAI API and the context length exceeds your organizations allowed tokens per minute for the specific model, you will receive the following error:

Rate limit reached for default-gpt-4-large in organization org-xxxxxxxxxxxx on tokens per min. 

Limit: 150000 / min. Current: 0 / min. Contact us through our help center at help.openai.com if you continue to have issues.

This is confusing because it makes the issue appear to be at the organization level, when in fact a bug in your code is just causing your context to be too large. Instead, the following error message should appear:

This model's maximum context length is 4097 tokens, however you requested 150678 tokens (150000 in your prompt; 678 for the completion). Please reduce your prompt; or completion length."

To Reproduce

1. Use the OpenAI python client
2. Use any open ai chat completions endpoint with far too large a context length
3. Observe error

Code snippets

No response

OS

macOS

Python version

^3.9

Library version

^0.27.8

[perryHV]

Hii @asim-shrestha I want to work on this issue please assign me.

[asim-shrestha]

@perryHV I don't have the auth to do that :) I imagine you can just make a PR

[ansumanswain]

could you attach the corresponding repo?

[asim-shrestha]

Assuming it should all be in this repo

[ansumanswain]

yes, Assuming it should all be in this repo

[rattrayalex]

This is backend API behavior, not Python library behavior.

I do agree that messaging could be better, so I hope we'll be able to make an associated backend change.

[asim-shrestha]

Awesome, Stainless looks great by the way