bug: o4-mini is not in the list of available models; yet it is the default

You do not seem to have access to o4-mini or o3 yet, and therefore these do not show up in the list of models in the dropdown. In this case we should not default to o4-mini.

To answer your other questions, yes this project is very much alive and under active development, please read through the documentation to get a sense of how to use it and what to expect!

Thibault,
Thank you for the reply.
Apparently the /model command produces a long list of models (including 4.1
). Th3 non obvious thing - the list has no visual indication that there are
models above the current selection. Major 🤦‍♀️

Pass that, if I select "o4-mini", I'm getting the invalid_request_error -
your organization must be verified.
With he link to provide all sorts of personal info...

On Mon, Apr 21, 2025, 2:49 PM Thibault Sottiaux @.***>
wrote:

You do not seem to have access to o4-mini or o3 yet, and therefore these do not show up in the list of models in the dropdown. In this case we should not default to o4-mini. To answer your other questions, yes this project is very much alive and under active development, please read through the documentation to get a sense of how to use it and what to expect! — Reply to this email directly, view it on GitHub <https://github.com/openai/codex/issues/486#issuecomment-2819246246>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABGHHXKQYKI4YXLT7XWZSWT22U4S5AVCNFSM6AAAAAB3RZR6UOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJZGI2DMMRUGY> . You are receiving this because you authored the thread.Message ID: @.***> tibo-openai left a comment (openai/codex#486) <https://github.com/openai/codex/issues/486#issuecomment-2819246246> You do not seem to have access to o4-mini or o3 yet, and therefore these do not show up in the list of models in the dropdown. In this case we should not default to o4-mini. To answer your other questions, yes this project is very much alive and under active development, please read through the documentation to get a sense of how to use it and what to expect! — Reply to this email directly, view it on GitHub <https://github.com/openai/codex/issues/486#issuecomment-2819246246>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABGHHXKQYKI4YXLT7XWZSWT22U4S5AVCNFSM6AAAAAB3RZR6UOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJZGI2DMMRUGY> . You are receiving this because you authored the thread.Message ID: @.***>

I'm sorry to complain, but the first-time user experience is significantly worse compared to Anthrope's product.

Thibault, Thank you for the reply. Apparently the /model command produces a long list of models (including 4.1 ). Th3 non obvious thing - the list has no visual indication that there are models above the current selection. Major 🤦‍♀️ Pass that, if I select "o4-mini", I'm getting the invalid_request_error - your organization must be verified. With he link to provide all sorts of personal info... […](#)

Looks like there's an extra step we have to take to access this via the API key.
I got the same error and it tells us to go to 'https://platform.openai.com/settings/organization/general'
Once there, we have to 'verify' our organization by adding a credit card. 👀

If you go to Billing on the left, the center context points out that your org is on a "Free Trail", and that 'This does not reflect the status of your ChatGPT account.'

So maybe with ChatGPT Plus you can only access the models via the webapp and not the API?

Sorry I can't test further, in case the API calls are more expensive than I can handle! 😅

🤣
Confirmed.
When I select one of the available models I get this error:
You exceeded your current quota, please check your plan and billing details

Weak sauce!

> Thibault, > Thank you for the reply. > Apparently the /model command produces a long list of models (including 4.1 > ). Th3 non obvious thing - the list has no visual indication that there are > models above the current selection. Major 🤦‍♀️ > Pass that, if I select "o4-mini", I'm getting the invalid_request_error - > your organization must be verified. > With he link to provide all sorts of personal info... > […](#) Looks like there's an extra step we have to take to access this via the API key. I got the same error and it tells us to go to 'https://platform.openai.com/settings/organization/general' Once there, we have to 'verify' our organization by adding a credit card. 👀 If you go to Billing on the left, the center context points out that your org is on a "Free Trail", and that 'This does not reflect the status of your ChatGPT account.' So maybe with ChatGPT Plus you can only access the models via the webapp and not the API? Sorry I can't test further, in case the API calls are more expensive than I can handle! 😅

not sure... I get :

!Image

clearly I'm so _not_ doing any of that to just access some models.

At the same time, i do _not_ see any "trial" messages. I used my business credit card to purchase credits.. It does say "Pay as you go" plan :

!Image

I'd suggest trying gpt-4.1 which shouldn't require verification. For using reasoning models in the API, developer verification is required for now at time of writing. Sorry for the confusion and inconvenience @az-z — we definitely want to make the onboarding experience smoother for first time users, thanks for giving it a try!

It's very unfortunate that the README of this project doesn't mention that biometric verification is required in order to use it with any of the recommended models (codex-mini-latest or o4-mini). I definitely wouldn't have put $20 into my OpenAI account if I had known it, the README is crazily misleading.