/review command ignores configured gpt-oss:120b model and tries to use gpt-5.1-codex instead

What version of Codex is running?

codex-cli 0.63.0

What subscription do you have?

pro

Which model were you using?

gpt-oss:120b

What platform is your computer?

Linux 6.2.0-generic x86_64 x86_64

What issue are you seeing?

When using the /review slash command, Codex ignores my configured model gpt-oss:120b (served on a local GPU server in my local network) and instead tries to use the gpt-5.1-codex model.

In other words:

Normal usage is configured to talk to a locally served gpt-oss:120b model.

As soon as I trigger /review and then select any of the submenus/options, Codex attempts to call gpt-5.1-codex instead of gpt-oss:120b.

This appears to be specific to the /review slash command behavior: it seems to bypass my configured model and force gpt-5.1-codex.

What steps can reproduce the bug?

1. Configure Codex CLI to use a locally served model:

Model: gpt-oss:120b
Endpoint: local GPU server on the same network (self-hosted with OLlama)

1. Start Codex in a project directory where the /review slash command is available.
2. Type /review in the Codex prompt.
3. In the /review menu, select any of the submenus/options.
4. Observe which model Codex tries to use.

What is the expected behavior?

Codex should respect the configured model and continue using gpt-oss:120b (the locally served model) when running /review.

All slash commands, including /review, should use the same model configuration that is applied to normal Codex interactions.

Additional information

_No response_