API Rate Limits Make Codex Unusable

What version of Codex is running?

Newest

Which model were you using?

Codex Medium

What platform is your computer?

_No response_

What steps can reproduce the bug?

Simply run Codex over the API with Level 1 or Level 2 api limits.

What is the expected behavior?

I dont want to get rate limited for normal usage.

What do you see instead?

_No response_

Additional information

I understand the reason for TPM rate limits over the API to prevent abuse, but the current limits are simply far too low for Codex usage to use it over the API. Even at Tier 2, I am hitting them once per minute, effectively making it useless. Codex uses a lot more tokens that your traditional chatbot app, and the TPM should reflect that difference for all Codex models.

There's no point in even using it like that. Rather than switch to API when my normal plan hits usage limits, now I just have to sit on the sidelines or use another provider. That's a terrible business model that literally pushes me to spend money with competing labs.

Please increase it. I can't even increase my level to level 3 because the rate limits make using it completely pointless.

<img width="1583" height="456" alt="Image" src="https://github.com/user-attachments/assets/572906cd-cb59-4476-a6d7-f78a709564ee" />

This happens OVER and OVER and OVER, and I'm not doing anything special. Just coding up a little ACP adapter.

TLDR; rate limits are for abuse, not normal usage. This arbitrary rate limit is so restrictive, it makes your product unusable.