model choosing error

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

26.818.41509

What subscription do you have?

ChatGPT Plus

What platform is your computer?

Darwin 25.4.0 arm64 arm

What issue are you seeing?

I selected GPT-5.6 in the Codex App, but when I asked the assistant which model it was using, it repeatedly reported a different model (GPT-5.5-mini).

During the same conversation, the assistant gave multiple inconsistent answers about its model identity, including GPT-5.5-mini, GPT-5.5 Thinking, and GPT-5.6 Sol.

The model selected in the UI was GPT-5.6, so I am unable to determine whether:

1. the app routed the conversation to the wrong backend model, or
2. the correct model was used but the assistant received or reported incorrect model identity information.

Expected behavior:
When GPT-5.6 is explicitly selected in the model selector, the request should be routed to GPT-5.6, and the assistant's reported model identity should be consistent with the selected/routed model.

Please investigate whether there is a mismatch between the Codex App model selector, backend model routing, and the model identity information exposed to the assistant.

What steps can reproduce the bug?

Feedback ID: no-active-thread-01a02901-da06-7a23-85af-e7439db1db3e

What is the expected behavior?

Please give me feedback and tell me why the condition is happening..

Additional information

No response

No source file or test is named. Start by tracing the Codex App model selector and the backend routing or model-identity path, using feedback ID no-active-thread-01a02901-da06-7a23-85af-e7439db1db3e. Determine whether GPT-5.6 was routed correctly or whether the identity exposed to the assistant is wrong, and verify that the selected and reported models are consistent.