Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| const [countries, setCountries] = useState<string[]>([]); | ||
| const [title, setTitle] = useState(''); | ||
| const [isManualMode, setIsManualMode] = useState(false); | ||
| const [isManualMode, setIsManualMode] = useState(true); |
There was a problem hiding this comment.
The new default sets isManualMode to true, so the UI now shows the VLM toggle off by default. However, the multi-image upload path still always appends model_name from selectedVlmModel before posting to /api/images/multi, so the backend will invoke a VLM even when the toggle is off. Users uploading multiple files without re-enabling the toggle will unexpectedly get AI-generated captions instead of staying in manual mode. Consider skipping the model_name for multi uploads when manual mode is selected so the default matches behavior.
Useful? React with 👍 / 👎.
1 participant
No description provided.