Codex app sends empty model ("") with ChatGPT account, causes unsupported model error

Resolved 💬 1 comment Opened Feb 4, 2026 by bluey-user Closed Feb 4, 2026

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

Version 260203.1501 (523)

What subscription do you have?

ChatGPT subscriptions Enterprise

What issue are you seeing?

When using Codex with a ChatGPT account, prompts fail and the UI renders a raw backend error:

{"detail":"The '' model is not supported when using Codex with a ChatGPT account."}

It appears Codex is sending an empty model name ("") instead of a supported GPT model.
The error is repeated on every prompt and is shown as assistant output rather than a UI error.

What steps can reproduce the bug?

  1. Sign in to the Codex app using a ChatGPT account
  2. Trigger an automation (any automation that prepares context or output)
  3. Wait until the automation completes and Codex indicates it is ready
  4. Ask a follow-up question related to the automation output

(e.g., "can you summarize the changes made")

  1. Observe that instead of an assistant response, Codex returns:

{"detail":"The '' model is not supported when using Codex with a ChatGPT account."}

  1. Repeat any follow-up question and observe the same error rendered as chat output each time

What is the expected behavior?

The system should either return a valid response to the follow-up query or present a user-friendly error message when the request cannot be fulfilled.

Additional information

_No response_

View original on GitHub ↗

This issue has 1 comment on GitHub. Read the full discussion on GitHub ↗