openai / openai/codex

[Windows] Reasoning effort locks to Instant after 50-message weekly Pro Chat allowance is exhausted; web/mobile unaffected

Open
#44,664 2 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

app bug rate-limits windows-os
Dominant language
Rust
Stars
125k
Forks
19.4k
PR merge metrics
PR metrics pending

Description

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

ChatGPT for Windows Version 26.903.61454 Release date: September 9, 2026 Powered by Codex & OWL

What subscription do you have?

GPT PRO

What platform is your computer?

Microsoft Windows NT 10.0.26200.0 x64

What issue are you seeing?

After exhausting the 50-message weekly Pro-model allowance in Chat, the reasoning effort selector in the Windows desktop app became locked to Instant.

This does not appear to be related to my Codex usage allowance.

My Codex usage allowance had already been exhausted well before this problem started. Even while my Codex allowance was exhausted, the Windows desktop app continued to allow me to select other reasoning efforts normally, and I did not experience this issue.

The problem started specifically after I used all 50 messages of my weekly Pro-model Chat allowance.

Timeline
  1. My Codex/Work usage allowance was exhausted.
  2. I continued using Chat normally.
  3. Medium, High, and Extra High reasoning efforts continued to work normally in the Windows desktop app.
  4. Later, I exhausted all 50 messages of my weekly Pro-model Chat allowance.
  5. Immediately after the Pro-model Chat allowance was exhausted, the Windows desktop app began forcing the reasoning effort to Instant.
  6. Medium, High, and Extra High could no longer be selected/applied in the Windows desktop app.

However, using the same account and the same exhausted Pro-model allowance state:

  • chatgpt.com still allows me to select and use the available non-Pro reasoning levels.
  • The mobile ChatGPT app also allows the available reasoning levels to be selected normally.
  • Only the Windows desktop app is locked to Instant.

This is important because the Chat Pro-model allowance and Codex/Work allowance are separate. My exhausted Codex allowance did not trigger this behavior. The behavior only appeared when the separate 50-message weekly Pro-model Chat allowance reached zero.

This suggests that the Windows desktop app may be incorrectly interpreting an exhausted Pro-model Chat allowance as a restriction on the entire reasoning-effort selector, rather than only disabling the Pro model(s) whose allowance has been exhausted.

There is a related Windows issue, #42435, where the reasoning effort resets from Extra High to Instant while the same account works correctly on the web. However, in my case I have identified a specific trigger: exhausting the weekly 50-message Pro-model Chat allowance.

What steps can reproduce the bug?

Feedback ID: no-active-thread-01a08d76-877d-7c71-8949-5be69f16c32d

What is the expected behavior?

No response

Additional information

No response

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.

Research direction

Begin by reproducing the Windows desktop behavior after the weekly Pro-model Chat allowance is exhausted, comparing it with chatgpt.com and the mobile app. Review related issue #42435 for context; done means available non-Pro reasoning levels remain selectable instead of the selector being forced to Instant.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust
Domain
desktop, operating-systems
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.