[Bug]: cannot change text generation model within the codex provider

[nathanscain]

Before submitting

• I searched existing issues and did not find a duplicate.
• I included enough detail to reproduce or investigate the problem.

Area

apps/desktop

Steps to reproduce

1. Go to settings
2. Attempt to change from 5.4 Mini High to anything else in codex
3. Witness it quickly revert back to 5.4 Mini High

Expected behavior

Should change to the selected model

Actual behavior

quickly reverts back to 5.4 Mini High after any change (but anthropic models can be freely selected)

Impact

Minor bug or occasional failure

Version or commit

0.0.15

Environment

macOS 26.3

Logs or stack traces

Screenshots, recordings, or supporting files

t3code-text-generation-model-selector-bug.mp4

Workaround

No response

[empotts]

also seeing this

[Perryman]

Confirming this on a fresh install of the desktop app (downloaded from t3.codes), no threads created yet. Same behavior: selecting any Codex/OpenAI model or changing the thinking level under Settings → Text generation model snaps back to gpt5.4 mini / thinking high after about a second. Claude models persist without issue.

Version: 0.0.15
Environment: macOS 26.3.1 (a), MacBook Pro M5, desktop app