docs: update model references to latest versions#608
docs: update model references to latest versions#608factory-ben wants to merge 1 commit intomainfrom
Conversation
Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
![factory-droid[bot]](https://avatars.githubusercontent.com/in/358017?s=60&v=4){kind=small}
| @@ -101,14 +101,11 @@ droid exec --auto high "Run tests, commit, and push changes" | |||
| | :---------------------------- | :--------------------------- | :-------------------------------- | :---------------- | | |||
| | `claude-opus-4-5-20251101` | Claude Opus 4.5 (default) | Yes (Off/Low/Medium/High) | off | | |||
| | `gpt-5.1-codex-max` | GPT-5.1-Codex-Max | Yes (Low/Medium/High/Extra High) | medium | | |||
There was a problem hiding this comment.
[P1] Don’t list unsupported reasoning level None for gpt-5.2-codex
This table says gpt-5.2-codex supports None, but elsewhere in these docs you describe reasoning levels as off|low|medium|high; if None isn’t actually accepted by droid exec, this will cause immediate CLI errors for users copying the command/setting.
| - **`low`**, **`medium`**, **`high`** – progressively increase deliberation time for more complex reasoning. | ||
|
|
||
| Anthropic models default to `off`, while GPT-5 starts on `medium`. | ||
| Anthropic models default to `off`, while GPT-5.2 starts on `low`. |
There was a problem hiding this comment.
[P1] Don’t claim GPT-5.2 defaults to low reasoning without verifying
This changes the stated default from medium to low; if the actual CLI/model default is still medium (or model-dependent), this is user-facing incorrect behavior guidance that will lead to confusing mismatches when people follow the docs.
There was a problem hiding this comment.
Updates look generally consistent, but there are a couple places where the docs now assert specific reasoning-level support/defaults that may not match the CLI’s accepted values. If those are inaccurate, users will hit immediate errors or see behavior that contradicts the documentation.
1 participant
No description provided.