Critical UX Violations in Model Selection Pop-up: Glaring Terminology Mismatches & Severe Workflow Friction

Open
#35,767 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
48/100
Issue type
Bug
Clarity
Mostly clear
Activity status
Quiet
Tech stack
macos
Domain
design, desktop

Research direction

Start by reproducing the workflow in the Codex desktop application's model selection pop-up, beginning from a new or existing session and opening Advanced settings. Done means model selection and reasoning effort are accessible from the main chat interface within two clicks, terminology is consistent, and the discrete choices use an immediate-response control without unnecessary animation.

Written by the indexing model from the issue text.

Description

app enhancement
What version of the Codex App are you using (From “About Codex” dialog)?

Version 26.721.41059

Released Jul 25, 2026

What subscription do you have?

plus

What platform is your computer?

macos

What issue are you seeing?

Hidden Model Selection: The model selection menu is buried under "Advanced" settings. Industry standard is a clear, 2-step max dropdown. Hiding this adds unnecessary friction for power users.

Inconsistent Terminology: The labels for model intensity/reasoning levels are wildly inconsistent (e.g., mixing "Light" with "Minimal", and "Max" with "Extra High").

Unnecessary UI Elements: The model intensity selection uses a slider/progress bar with unnecessary animations that add zero functional value.

What steps can reproduce the bug?

Open the Codex desktop application and start a new session or open an existing one.

Attempt to change the active model or adjust the reasoning effort for the current prompt.

Observe that the model selection is not immediately accessible and requires digging into the "Advanced" settings menu.

Once inside the pop-up, observe the UI components: an animated progress bar/slider is used for discrete model intensity selection, and the terminology is completely mismatched (e.g., UI labels showing "Light" and "Max" while the tooltips or other UI states show "Minimal" and "Extra High").

What is the expected behavior?

Immediate Access: Model selection and reasoning effort should be available directly on the main chat interface (e.g., via a clear dropdown or segmented control), requiring no more than 2 clicks. It should absolutely not be hidden under "Advanced" settings.

Consistent Terminology: Terms used for model intensity must be unified across the entire application (choose either Light/Max OR Minimal/Extra High, and stick to it).

Functional UI over "Clever" Design: Replace the animated slider/progress bar with standard, instant-response UI components. Users need clarity and speed, not unnecessary visual animations for a settings toggle.

Additional information

No response

Dominant language
Rust
Stars
125k
Forks
19.5k
Avg merge
1m
Merged PRs (30d)
1k

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from openai/codex

All issues in openai/codex

Similar issues

More Rust issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.