Codex Desktop: "Selected model is at capacity" for all models on one Plus account while another Plus account works normally #44113

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

152.0.7977.83

What subscription do you have?

ChatGPT Plus

What platform is your computer?

Darwin 25.2.0 arm64 arm

What issue are you seeing?

Issue Description

I am experiencing a persistent issue in Codex Desktop App.

The error message is:

"Selected model is at capacity. Please try a different model."

This issue occurs repeatedly during normal Codex usage.

Environment

• Codex App version: 152.0.7977.83
• Subscription: ChatGPT Plus
• Feedback ID: 01a07fb8-ace3-7c30-a8fd-79d41781f936

Problem Details

The issue is not limited to a specific model.

I have tested multiple available models in Codex, and all of them return the same capacity error.

Reproduction Test

I compared two ChatGPT Plus accounts under the exact same environment:

Affected account:

• Same computer
• Same network
• Same Codex App version
• Same project
• Same models

Result:
All models fail with:
"Selected model is at capacity. Please try a different model."

Another Plus account:

• Same computer
• Same network
• Same Codex App version
• Same project
• Same models

Result:
Works normally.

Impact

This issue significantly affects my workflow:

• Long coding tasks are interrupted repeatedly.
• Retrying failed requests consumes additional tokens.
• Interrupted tasks lose context continuity.
• Response quality decreases after repeated retries.

Expected Behavior

Codex should allow normal model usage with my Plus account.

Actual Behavior

All available models fail with capac

ity errors only on this account.

Additional Information

I have already contacted OpenAI Support and the case has been escalated.

Since another Plus account works normally under the same environment, this appears potentially related to:

• account-specific backend routing
• account configuration
• capacity allocation
• account-level throttling

Could the team please investigate?

Thank you.

What steps can reproduce the bug?

Steps to reproduce

1. Open Codex Desktop App.

2. Sign in with the affected ChatGPT Plus account.

3. Open any project.

4. Select any available Codex model.

5. Send a normal coding request.

6. The request fails with:

"Selected model is at capacity. Please try a different model."

7. Switch to another available model and retry.

8. The same error occurs again.

Additional reproduction information

The issue can be reproduced consistently on my account.

I also performed an account comparison test:

Affected account:

• Same computer
• Same network
• Same Codex App version (152.0.7977.83)
• Same project
• Same models

Result:
All models fail.

Another ChatGPT Plus account:

• Same computer
• Same network
• Same Codex App version
• Same project
• Same models

Result:
Works normally.

Feedback ID:
01a07fb8-ace3-7c30-a8fd-79d41781f936

What is the expected behavior?

Codex should successfully process requests using available models for my ChatGPT Plus account.

Model availability should behave consistently with other Plus accounts under the same environment.

Long-running coding tasks should complete without repeated capacity failures or interruptions.

Additional information

Additional information

This issue has already been reported to OpenAI Support and escalated to a support specialist.

Support case details:

• Feedback ID:
  01a07fb8-ace3-7c30-a8fd-79d41781f936

OpenAI Support has confirmed that:

• No additional reinstall/network/project/model troubleshooting is required at this stage.
• The comparison test between two Plus accounts provided useful isolation information.

The issue is significantly affecting my workflow:

• A single coding task can be interrupted multiple times.
• Retrying requests consumes additional tokens.
• Interrupted sessions lose context continuity.
• Response quality can degrade after repeated retries.

I was also considering upgrading from ChatGPT Plus to Pro because of my Codex usage, but I am currently waiting for this account-specific issue to be resolved before upgrading.

This appears potentially related to:

• account-specific backend routing
• account configuration
• capacity allocation
• account-level throttling

Any investigation or guidance from the Codex team would be greatly appreciated.