OfflineLM

A private, dependency-free chat interface for Chrome's built-in LanguageModel API.

Run locally

Service workers and Chrome's built-in AI require a secure context. Serve this folder from localhost instead of opening index.html directly. For example:

python -m http.server 8080

Then open http://localhost:8080 in a supported desktop version of Chrome.

OfflineLM checks LanguageModel.availability(), presents Chrome's one-time model download with progress, and only enables chat after the model is ready. Conversations are stored in localStorage and can be exported or imported from Settings.