Can't use some models

Resolved 💬 5 comments Opened Apr 17, 2025 by notlikejuice Closed May 17, 2025

💡 Likely answer: A maintainer (tibo-openai, collaborator) responded on this thread — see the highlighted reply below.

Logs:

│ ↳ model: o4-mini                                             │
│ ↳ approval: suggest                                          │
╰──────────────────────────────────────────────────────────────╯
user
test

    system
    ⚠️  OpenAI rejected the request (request ID: req_1800747adab1d5d22256b202de8e8542). Please verify your settings and
    try again.

    system
    Switched model to o4-mini

    system
    Switched model to chatgpt-4o-latest
user
test

    system
    ⚠️  OpenAI rejected the request (request ID: req_75fb5da5cd5c27f0a02d365651a20dfc). Please verify your settings and
    try again.
╭──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ send a message                                                                                                       │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
  send q or ctrl+c to exit | send "/clear" to reset | send "/help" for commands | press enter to send

╭──────────────────────────────────────────────────────────────────────────────╮
│ Switch model                                                                 │
│ Current model: chatgpt-4o-latest                                             │
│                                                                              │
│                                                                              │

    system
    Switched model to chatgpt-4o-latest

    system
    Switched model to ft:gpt-4o-2024-08-06:pruszyn:my-exp4:AE8LHDHD:ckpt-step-80
user
test

    codex
    I'm here! How can I assist you today with your code?
╭──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ send a message                                                                                                       │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
  send q or ctrl+c to exit | send "/clear" to reset | send "/help" for commands | press enter to send

I am unable to the latest models, as well as custom ones such as chatgpt-4o-latest or o1-mini, among others. Only 4o and 4o-mini are functional, while 4.1 does not work either and of course o4-mini and o3.,

My api are:
Your organization is currently in Usage tier 3. Your limits will automatically be increased once you move to the next usage tier based on the criteria outlined below. Visit our usage tiers documentation to learn more about the limits associated with each tier.
Current tier
Usage tier 3
Once the following criteria are met, you'll automatically move to the next tier:
At least $250 spent on the API since account creation.
At least 14 days passed since first successful payment.
View payment history
Buy credits

View original on GitHub ↗

5 Comments

tibo-openai collaborator · 1 year ago

We're looking.

trevorcreech contributor · 1 year ago

@notlikejuice can you try again with the latest version of codex? You can update with npm install -g @openai/codex

alextes · 1 year ago

had the same issue with the cli as released right after the announcement i think. plus account. default o4-mini. old api token.
ran an npm upgrade check, didn't upgrade codex.
used the upgrade command as suggested by @trevorcreech anyway.
fixed! (using the default o4-mini) i think this can be closed.

perhaps nice to update the title to "codex fails with 'OpenAI rejected the request' with default 4o-mini model"

notlikejuice · 1 year ago

@notlikejuice can you try again with the latest version of codex? You can update with npm install -g @openai/codex

0.1.2505161800
anthropic/claude-3.5-haiku-20241022:beta
linux | x64 | 5.15.167.4-microsoft-standard-WSL2

- ```
  test
  
  - `0 reasoning` | `0 tool`
- 
  test
  
  - `0 reasoning` | `0 tool`
- 
  test

  - `0 reasoning` | `0 tool`

When I enter the default “codex” command and enable it with openai, the models from the original provider work naturally, but when I change it via /models -> tab -> model as above, there is no message. Problems start to occur when I enable codex with the command “codex --provider openrouter”.

0.1.2505161800
openai/chatgpt-4o-latest
linux | x64 | 5.15.167.4-microsoft-standard-WSL2

codex --provider openrouter                                                                             
╭──────────────────────────────────────────────────────────────╮
│ ● OpenAI Codex (research preview) v0.1.2505161800            │
╰──────────────────────────────────────────────────────────────╯
╭──────────────────────────────────────────────────────────────╮
│ localhost session: 9263d86ccf8f4b68b6e35907b8f17812          │
│ ↳ workdir: /mnt/c/Users/karol/codexxx                        │
│ ↳ model: anthropic/claude-3.5-haiku-20241022:beta            │
│ ↳ provider: openrouter                                       │
│ ↳ approval: suggest                                          │
╰──────────────────────────────────────────────────────────────╯
user
test

    system
    ⚠️  OpenAI rejected the request. Error details: Status: 401, Code: 401, Type: unknown, Message: 401 ╯
    No auth credentials found. Please verify your settings and try again.

    system
    Switched model to agentica-org/deepcoder-14b-preview:free
user
haha

    system
    ⚠️  OpenAI rejected the request. Error details: Status: 401, Code: 401, Type: unknown, Message: 401 ╯
    No auth credentials found. Please verify your settings and try again.

    system
    Switched model to openai/chatgpt-4o-latest
user
test

    system
    ⚠️  OpenAI rejected the request. Error details: Status: 401, Code: 401, Type: unknown, Message: 401 ╯
    No auth credentials found. Please verify your settings and try again.

    system
    🔗 Bug report URL: https://github.com/openai/codex/issues/new?template=2-bug-report.yml&labels=bug&v
    ersion=0.1.2505161800&model=openai%2Fchatgpt-4o-latest&platform=%60linux%60+%7C+%60x64%60+%7C+%605.1
    5.167.4-microsoft-standard-WSL2%60&steps=-+%60%60%60%0A++test%0A++%60%60%60%0A++-+%600+reasoning%60+
    %7C+%600+tool%60%0A-+%60%60%60%0A++haha%0A++%60%60%60%0A++-+%600+reasoning%60+%7C+%600+tool%60%0A-+%
    60%60%60%0A++test%0A++%60%60%60%0A++-+%600+reasoning%60+%7C+%600+tool%60

    system
    🔗 Bug report URL: https://github.com/openai/codex/issues/new?template=2-bug-report.yml&labels=bug&v
    ersion=0.1.2505161800&model=openai%2Fchatgpt-4o-latest&platform=%60linux%60+%7C+%60x64%60+%7C+%605.1
    5.167.4-microsoft-standard-WSL2%60&steps=-+%60%60%60%0A++test%0A++%60%60%60%0A++-+%600+reasoning%60+
    %7C+%600+tool%60%0A-+%60%60%60%0A++haha%0A++%60%60%60%0A++-+%600+reasoning%60+%7C+%600+tool%60%0A-+%
    60%60%60%0A++test%0A++%60%60%60%0A++-+%600+reasoning%60+%7C+%600+tool%60

    system
    Invalid command "/m". Use /help to retrieve the list of commands.

    system
    Switched provider to OpenAI with model openai/chatgpt-4o-latest

    system
    Switched model to codex-mini-latest
user
test

    system
    ⚠️  OpenAI rejected the request (request ID: req_1c5177da82e4c145b832300658a50976). Error details:  ╯
    Status: 404, Code: unknown, Type: invalid_request_error, Message: 404 This model is only supported  
    in v1/responses and not in v1/chat/completions.. Please verify your settings and try again.
╭──────────────────────────────────────────────────────────────────────────────────────────────────────╮
│                                                                                                      │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────╯
  ctrl+c to exit | "/" to see commands | enter to send — 100% context left

Models on codex --provider openrouter like an openai/models doesn't working, either clicking with /models -> tab -> change for openai doesn't helps.

notlikejuice · 1 year ago

had the same issue with the cli as released right after the announcement i think. plus account. default o4-mini. old api token. ran an npm upgrade check, didn't upgrade codex. used the upgrade command as suggested by @trevorcreech anyway. fixed! (using the default o4-mini) i think this can be closed. perhaps nice to update the title to "codex fails with 'OpenAI rejected the request' with default 4o-mini model"

Sure, could i open a new issue with others models i think.