Non o4-mini models all fail

What version of Codex is running?

v0.1.2504172351

Which model were you using?

gpt-4.1, o3-mini, gpt-4o

What platform is your computer?

Windows WSL2 Ubuntu 24.04

What steps can reproduce the bug?

When using any other model than o4-mini, Codex always fails. It seems this system wasn't really ready for release with all the issues going on, like not implementing cooldown feature knowing OpenAI implements a tiered rate limit. Seeking out using other models reveals they cannot properly use Codex.

What is the expected behavior?

Competent outputs from available models to use.

What do you see instead?

Various errors. For example GPT-4o will say

Error: spawn cat /home/was/projects/codex/yeti_toolkit/yeti/sdk/io.py ENOENT

codex It seems we're having persistent issues accessing the file. To move forward, could you try providing the exact

content or confirming whether the path is accessible from your current working environment? This would help further
debug and address the syntax issue.

The path is accessible, file is there, and previously edited by another model. Similarly GPT 4.1 will says

Error: spawn cat /home/was/projects/codex/yeti_toolkit/yeti/sdk/io.py ENOENT

codex

It appears there's a problem with accessing the file. Could you check if the file path is correct or if there are
permissions issues? Alternatively, let me know if there might be another way to access or verify the file content.

I'll also often get the following with all models besides o4-mini:

[Error]: An error occurred while processing your request. You can retry your request, or contact us through our help center at help.openai.com if the error persists. Please include the request ID req_88bcd72c04130b281b463d7e80f356e9 in your message.
    at e.a [as iterator] (file:///home/was/.nvm/versions/node/v22.14.0/lib/node_modules/@openai/codex/dist/cli.js:449:1514)
    at process.processTicksAndRejections (node:internal/process/task_queues:105:5)
    at async T0.run (file:///home/was/.nvm/versions/node/v22.14.0/lib/node_modules/@openai/codex/dist/cli.js:466:2933) {
  status: undefined,
  headers: undefined,
  request_id: undefined,
  error: {
    type: 'model_error',
    code: null,
    message: 'An error occurred while processing your request. You can retry your request, or contact us through our help center at help.openai.com if the error persists. Please include the request ID req_88bcd72c04130b281b463d7e80f356e9 in your message.',
    param: null
  },
  code: null,
  param: null,
  type: 'model_error'
}

Additional information

o4-mini operated just fine until this rate limit issue, now it's unusable, I guess until my tier is upgraded in 5 days per the wait limit, for an account already in good standing with OpenAI for years, just cause I want to use the API.

The rest of the models do not appear to properly work with Codex and can't get a single request to finish. This all should have been pretty rigorously tested prior to launch.

I've tried rolling back to working versions of the software from from github that it should have context on, and even unreleased version from our gitlab, these other models just aren't working with Codex.

So far I have chewed through 10 dollars, for no output...