gpt-5.5 shown in Codex, but rejected for ChatGPT Plus account with 400 unsupported-model error

What version of Codex CLI is running?

codex-cli 0.125.0

What subscription do you have?

ChatGPT Plus

Which model were you using?

gpt-5.5

What platform is your computer?

Windows via WSL2

What terminal emulator and version are you using (if applicable)?

Codex CLI in WSL2 terminal

What issue are you seeing?

Codex shows my account as Plus in /status, and gpt-5.5 is visible/selectable, but requests fail immediately with:

```text
{"type":"error","status":400,"error":{"type":"invalid_request_error","message":"The 'gpt-5.5' model is not supported when using Codex with a ChatGPT account."}}

What steps can reproduce the bug?

Codex shows my account as Plus in /status, and gpt-5.5 is visible/selectable, but requests fail immediately with:

```text
{"type":"error","status":400,"error":{"type":"invalid_request_error","message":"The 'gpt-5.5' model is not supported when using Codex with a ChatGPT account."}}

Relevant context:

• /status shows:
• OpenAI Codex (v0.125.0)
• Account: [redacted] (Plus)
• Model: gpt-5.4
• Session: 019dc8f0-8c9b-70a0-9147-d456244229e0
• I am signed in with a ChatGPT account, not an API key.
• The model picker offers gpt-5.5, but execution is rejected by the backend.

This looks like an entitlement / account-sync inconsistency:

• account tier appears to be Plus
• but gpt-5.5 is still rejected as unsupported for ChatGPT account auth

What is the expected behavior?

1. Sign in to Codex CLI with a ChatGPT Plus account.
2. Confirm in /status that the account is shown as Plus.
3. Select gpt-5.5 in the model picker.
4. Send a simple prompt such as: hi
5. The request fails immediately with:

{"type":"error","status":400,"error":{"type":"invalid_request_error","message":"The 'gpt-5.5' model is not supported when using Codex with a ChatGPT account."}}

Thread id:
019dc8f0-8c9b-70a0-9147-d456244229e0

Additional information

If gpt-5.5 is shown as selectable for a ChatGPT Plus account, requests should run successfully.

At minimum, the model picker and backend entitlements should be consistent:

• either gpt-5.5 should not be shown
• or it should be executable for the signed-in Plus account