Openrouter api support #14

Activity

1. geogod42 commented on Apr 16, 2025

   

   geogod42

   on Apr 16, 2025

   More actions

   give me a hour

   👍React with 👍18Reacted by Carlos, Elliot Dohm, Toven, nitianxue, Juiyang Kuan, alang, Duino, Mick Clarke, SDGLBL, Peter Cunha and 8 more

   😄React with 😄7Reacted by Clay Mullis, Yukai Huang, chris wall, ChihayaAnno, 栗悟饭とカメハメ波, MikeUdin and Manan Santoki

   👀React with 👀4Reacted by Nannan A, Kenwwww, Riley Usagi and pavel444-byte
2. DiegoSpinola commented on Apr 16, 2025

   

   DiegoSpinola

   on Apr 16, 2025

   More actions

   you can use the env var OPENAI_BASE_URL

   ❤️React with ❤️16Reacted by Nannan A, Toven, Alex Chapin, lmdev, Tim Jünemann, Mp3Pintyo, Yukai Huang, Juiyang Kuan, chris wall, Patryk Zoltowski and 6 more
3. KorigamiK commented on Apr 16, 2025

   

   KorigamiK

   on Apr 16, 2025

   Last edited by KorigamiK

   More actions

   you can use the env var OPENAI_BASE_URL

   This doesn't seem to work right now even if you set up the baseurl and model names correctly since the agent loop requests the chat completions using the new openai /v1/responses api instead of the older /v1/chat/completions api which other providers already support.

   This however seems to be a deliberate way to discourage using your own models to run codex and some work on this front would be very helpful for all.

   👍React with 👍15Reacted by kmehtarbx, Shaun Pedicini, Yukai Huang, Diego Spinola, num13ru, Juiyang Kuan, 见微, Shane Sewell, Huge, Stone and 5 more
4. DiegoSpinola commented on Apr 16, 2025

   

   DiegoSpinola

   on Apr 16, 2025

   More actions

   So maybe adding something like this proxy to the mix:

   https://github.com/open-responses/open-responses

   👍React with 👍7Reacted by Zyad Alalgui, Juiyang Kuan, Duino, Hugo SANSON, Glenn 'devalias' Grant, Михаил Назаров and Peng-YM
5. tibo-openai

   added

   enhancementNew feature or requestNew feature or request

   on Apr 16, 2025
6. codefromthecrypt commented on Apr 16, 2025

   

   codefromthecrypt

   on Apr 16, 2025

   More actions

   Should we keep this titled as OpenRouter API support, or /chat/completions instead? If the latter, it basically opens up a floodgate of options.

   👍React with 👍6Reacted by Rick , Hugo SANSON, SDGLBL, 见微, Glenn 'devalias' Grant and Lukas
7. DD1302 commented on Apr 17, 2025

   

   DD1302

   on Apr 17, 2025

   More actions

   Is anyone still working on this? Would love to use openrouter on codex
8. Soren8 commented on Apr 18, 2025

   

   Soren8

   on Apr 18, 2025

   More actions

   Looks like there's a fork here https://github.com/ymichael/open-codex, but most discussion about this feature is in #26

   👍React with 👍7Reacted by Diego Spinola, Huge, Jan Seewald, theblazehen, Glenn 'devalias' Grant, Lukas and ejasmudar
9. codex-maintainers commented on Aug 7, 2025

   

   codex-maintainers

   on Aug 7, 2025

   More actions

   Thank you for the feedback! Please try again on the latest version of Codex CLI. If the issue persists, let us know.
10. codex-maintainers

    closed this as completedon Aug 7, 2025
11. timqian commented on Sep 8, 2025

    

    timqian

    on Sep 8, 2025

    More actions

    @codex-maintainers
    This feature does not work on the latest version of codex.

    > sudo npm i -g @openai/codex
    > codex --version
    codex-cli 0.30.0

    
12. timqian commented on Sep 8, 2025

    

    timqian

    on Sep 8, 2025

    More actions

    Found the solution, looks like the way to use other models has been changed, one needs to refer to this doc to use other models(by using config.toml): https://github.com/openai/codex/blob/main/docs/config.md

    This is a working config.toml to use openrouter if anyone need

    model_provider="openrouter"
    model="openai/gpt-5"
    
    [model_providers.openrouter]
    name="openrouter"
    base_url="https://openrouter.ai/api/v1"
    env_key="OPENROUTER_API_KEY"
    

    ❤️React with ❤️18Reacted by Glenn 'devalias' Grant, Yichang Jian, Gregory Petrosyan, Asif Mulla, Alexander Puga, hedeng, Muhammed İKİNCİ, ericshao, Rammiah, xuyangyuio and 8 more
13. zeekay

    added a commit that references this issue on Sep 8, 2025

    fix(tui/input): normalize KeyEvent kind on non-enhanced terminals to …

    ...

    This issue will close once commit 5b722e0 is merged into the 'main' branch.

    5b722e0
14. chongdashu commented on Sep 10, 2025

    

    chongdashu

    on Sep 10, 2025

    More actions

    Found the solution, looks like the way to use other models has been changed, one needs to refer to this doc to use other models(by using config.toml): https://github.com/openai/codex/blob/main/docs/config.md

    This is a working config.toml to use openrouter if anyone need

    model_provider="openrouter"
    model="openai/gpt-5"
    
    [model_providers.openrouter]
    name="openrouter"
    base_url="https://openrouter.ai/api/v1"
    env_key="OPENROUTER_API_KEY"
    

    This still doesn't work for me. After setting above, you you run codex -model openrouter ?
15. timqian commented on Sep 10, 2025

    

    timqian

    on Sep 10, 2025

    Last edited by timqian

    More actions

    Found the solution, looks like the way to use other models has been changed, one needs to refer to this doc to use other models(by using config.toml): https://github.com/openai/codex/blob/main/docs/config.md
    This is a working config.toml to use openrouter if anyone need

    model_provider="openrouter"
    model="openai/gpt-5"
    
    [model_providers.openrouter]
    name="openrouter"
    base_url="https://openrouter.ai/api/v1"
    env_key="OPENROUTER_API_KEY"
    

    This still doesn't work for me. After setting above, you you run codex -model openrouter ?

    after updating the config, just run codex works for me

    BTW, you will also need the OPENROUTER_API_KEY env in your .bashrc/ .zshrc

    👍React with 👍6Reacted by beyondkmp, Eason, bmbro, Asad Memon, Albert and Mohammad Banisaeid

    🎉React with 🎉2Reacted by Marcel Partap and Victor Henrique Grein
16. zcg

    added a commit that references this issue on Dec 4, 2025

    [Claude Code] After prompt openai#14

    c0c3e06
17. Galileo01 commented on Jun 3, 2026

    

    Galileo01

    on Jun 3, 2026

    Last edited by Galileo01

    More actions

    this config.toml ，It doesn't work for me，still need login

    codex-cli 0.136.0
18. gabrielbryk

    added 2 commits that reference this issue on Aug 3, 2026

    docs: retarget fork patch manifest at 0.147.0-alpha.4

    ...

    2f905bf

    docs: record upstream watch list and verification checklist

    ...

    c3b22c3
19. gabrielbryk

    added 4 commits that reference this issue on Aug 11, 2026

    docs: retarget fork patch manifest at 0.147.0-alpha.4

    ...

    fac7477

    docs: record upstream watch list and verification checklist

    ...

    3f1e05d

    docs: retarget fork patch manifest at 0.147.0-alpha.4

    ...

    7ab89a0

    docs: record upstream watch list and verification checklist

    ...

    4b22f65
20. jiangshengdev

    added a commit that references this issue on Sep 16, 2026

    feat(storybook): mark connection recovery simulation controls as DEV (o…

    ...

    3d5d662
21. friel-openai

    added a commit that references this issue on Sep 16, 2026

    Advance Codex to 0.151.0-alpha.7.2 (#14)

    ...

    bb1238c
22. soddygo

    added a commit that references this issue on Sep 24, 2026

    fix(bridges): first batch of confirmed review findings (P0 + 6 P1/P2)

    ...

    3d55420