Issue Station
Sign In
Tools hereToolbox

codex - 💡(How to fix) Fix Model selector is stuck on GPT-5.5 with medium intelligence in chats and some projects [2 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
openai/codex#19316•Fetched 2026-04-25 06:11:55
View on GitHub
Comments
2
Participants
2
Timeline
8
Reactions
0
Author
STSNaiveSTSNaive
Participants
github-actions[bot]github-actions[bot]STSNaiveSTSNaive
Timeline (top)
commented ×2labeled ×2closed ×1cross-referenced ×1

Root Cause

This is still disruptive because the model/intelligence choice usually needs to be made before sending the first prompt. For example, I may want to start a new conversation with a cheaper/faster model or a different reasoning level, but the UI initially only exposes gpt-5.5 with medium intelligence.

RAW_BUFFERClick to expand / collapse

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

Codex App 26.422.21637 (2056)

What subscription do you have?

ChatGPT Pro

What platform is your computer?

Darwin 25.4.0 arm64 arm

What issue are you seeing?

In the Codex App, the model selector appears to be restricted only in the initial/new-conversation state for Chats mode and for new conversations in some projects.

Before I send the first message in an affected chat/project conversation, the only available model/intelligence combination I can select is:

  • gpt-5.5
  • medium intelligence

I cannot switch to other models or other intelligence levels from the selector in that initial state, even though these options are expected to be available in the app.

After I send a message in the conversation, the selector appears to recover and I can choose other models/intelligence levels again.

This does not affect every project. So far I see it in Chats mode and when creating new conversations in some projects. Existing conversations, or conversations after the first message has been sent, may allow model selection normally.

This is still disruptive because the model/intelligence choice usually needs to be made before sending the first prompt. For example, I may want to start a new conversation with a cheaper/faster model or a different reasoning level, but the UI initially only exposes gpt-5.5 with medium intelligence.

Video placeholder:

https://github.com/user-attachments/assets/b814278b-cbb0-45bf-b07a-de4253b4da37

What steps can reproduce the bug?

  1. Open the Codex App.
  2. Go to Chats mode, or open one of the affected projects.
  3. Start a new conversation, but do not send any message yet.
  4. Click the model selector.
  5. Observe the available model/intelligence options.
  6. Try to select a model other than gpt-5.5, or an intelligence level other than medium.
  7. Send a first message in the conversation.
  8. Click the model selector again.

Observed result:

  • Before the first message is sent, the selector only allows gpt-5.5 with medium intelligence.
  • Other models/intelligence levels are not selectable from the UI in that initial state.
  • After the first message is sent, the selector appears to recover and other options become selectable again.
  • This happens in Chats mode and in some projects, but not necessarily every project.

What is the expected behavior?

The Codex App model selector should allow selecting the other models and intelligence levels available to my account/subscription before sending the first message in a new conversation.

Additional information

No response

extent analysis

TL;DR

The issue can be mitigated by sending a first message in the conversation to recover the model selector functionality.

Guidance

  • Verify that the issue only occurs in the initial state of new conversations in Chats mode and some projects.
  • Check if the model selector recovers after sending the first message in the conversation.
  • Try selecting a different model or intelligence level after sending the first message to confirm the selector is working as expected.
  • If the issue persists, consider testing in a different project or mode to isolate the problem.

Example

No code snippet is provided as the issue seems to be related to the Codex App's UI behavior.

Notes

The root cause of the issue is unclear, but it appears to be related to the initial state of new conversations. The provided steps to reproduce the bug can help in further debugging.

Recommendation

Apply workaround: Send a first message in the conversation to recover the model selector functionality, as this seems to temporarily resolve the issue.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#inference speed#output truncation#response parsing#generation error#database connection

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING