codex app local 404 Not Found: model 'gpt-5.5' not found, url: http://127.0.0.1:11434/v1/responses

well , i tried almost anything no still keep going back to same old story,
however if i open a new project i not have the issue. i am trying to
reinstate everything see if it help in any way , keep you updated tank you
for your info

Op do 28 mei 2026 om 08:25 schreef Sheikh Mohammad Ali Maheer <
@.***>:

maheerCodes left a comment (openai/codex#24795) <https://github.com/openai/codex/issues/24795#issuecomment-4559774665> This may be a model alias/version mismatch between the Codex app and the locally running inference bridge. The key detail is: 404 Not Found: model 'gpt-5.5' not found combined with: http://127.0.0.1:11434/v1/responses Port 11434 is commonly associated with local OpenAI-compatible runtimes, so the app is likely forwarding requests to a local provider that does not expose a model named exactly gpt-5.5. Possible causes: - the local backend only exposes different model IDs. - Codex app cached an old/default model mapping after update. - reconnect flow restores a stale model configuration. - local provider supports /v1/chat/completions but not /v1/responses . Would be useful to confirm: - what models are returned from: curl http://127.0.0.1:11434/v1/models - whether changing the configured model manually fixes reconnect. - whether deleting cached session/project state resolves it. Potential improvement for the app: instead of failing with a raw 404, detect unsupported model IDs and surface something clearer like: Configured model 'gpt-5.5' is unavailable from the connected local provider. That would make troubleshooting much easier for local-runtime users. — Reply to this email directly, view it on GitHub <https://github.com/openai/codex/issues/24795?email_source=notifications&email_token=CCHFO6ZIAW2JAF7UYB5V3M3446BQJA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINJVHE3TONBWGY22M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KYZTPN52GK4S7MNWGSY3L#issuecomment-4559774665>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/CCHFO62G5S5SHASEV76DC5D446BQJAVCNFSM6AAAAACZPTCW3WVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DKNJZG43TINRWGU> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/CCHFO66XXQFBQNQ2DRCHRG3446BQJA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINJVHE3TONBWGY22M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KUZTPN52GK4S7NFXXG> and Android <https://github.com/notifications/mobile/android/CCHFO6ZRGIM5LAAI2MCXAWT446BQJA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINJVHE3TONBWGY22M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2K4ZTPN52GK4S7MFXGI4TPNFSA>. Download it today! You are receiving this because you authored the thread.Message ID: @.***>

{

• "object": "list",
• "data":
• "id": "qwen3.6:latest", "object": "model", "created": 1779893084

, "owned_by": "library"

• "id": "gemini-3-flash-preview:latest", "object": "model", "created":

1779184316, "owned_by": "library"

• "id": "gemma4:latest", "object": "model", "created": 1778987612, "

owned_by": "library"

• "id": "Qwen3.5:latest", "object": "model", "created": 1778804364, "

owned_by": "library"

Op zo 31 mei 2026 om 10:07 schreef Olivier Vanlaer @.***>:

well , i tried almost anything no still keep going back to same old story, however if i open a new project i not have the issue. i am trying to reinstate everything see if it help in any way , keep you updated tank you for your info Op do 28 mei 2026 om 08:25 schreef Sheikh Mohammad Ali Maheer < @.***>: > maheerCodes left a comment (openai/codex#24795) > <https://github.com/openai/codex/issues/24795#issuecomment-4559774665> > > This may be a model alias/version mismatch between the Codex app and the > locally running inference bridge. > > The key detail is: > > 404 Not Found: model 'gpt-5.5' not found > > combined with: > > http://127.0.0.1:11434/v1/responses > > Port 11434 is commonly associated with local OpenAI-compatible runtimes, > so the app is likely forwarding requests to a local provider that does not > expose a model named exactly gpt-5.5. > > Possible causes: > > - the local backend only exposes different model IDs. > - Codex app cached an old/default model mapping after update. > - reconnect flow restores a stale model configuration. > - local provider supports /v1/chat/completions but not /v1/responses . > > Would be useful to confirm: > > - what models are returned from: > > curl http://127.0.0.1:11434/v1/models > > > - whether changing the configured model manually fixes reconnect. > - whether deleting cached session/project state resolves it. > > Potential improvement for the app: > instead of failing with a raw 404, detect unsupported model IDs and > surface something clearer like: > > Configured model 'gpt-5.5' is unavailable from the connected local provider. > > That would make troubleshooting much easier for local-runtime users. > > — > Reply to this email directly, view it on GitHub > <https://github.com/openai/codex/issues/24795?email_source=notifications&email_token=CCHFO6ZIAW2JAF7UYB5V3M3446BQJA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINJVHE3TONBWGY22M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KYZTPN52GK4S7MNWGSY3L#issuecomment-4559774665>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/CCHFO62G5S5SHASEV76DC5D446BQJAVCNFSM6AAAAACZPTCW3WVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DKNJZG43TINRWGU> > . > Triage notifications, keep track of coding agent tasks and review pull > requests on the go with GitHub Mobile for iOS > <https://github.com/notifications/mobile/ios/CCHFO66XXQFBQNQ2DRCHRG3446BQJA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINJVHE3TONBWGY22M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KUZTPN52GK4S7NFXXG> > and Android > <https://github.com/notifications/mobile/android/CCHFO6ZRGIM5LAAI2MCXAWT446BQJA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINJVHE3TONBWGY22M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2K4ZTPN52GK4S7MFXGI4TPNFSA>. > Download it today! > You are receiving this because you authored the thread.Message ID: > @.***> >