I built Apogee, an open-source browser extension that lets you summarize and ask questions about webpages, PDFs, DOCX files, videos, emails, and text using local AI models.
The goal is simple: keep your data on your device.
Apogee doesn't require a backend, account, or API key. It supports local inference through WebGPU/WebAssembly, with optional support for Ollama and llama.cpp.
The project was also selected for Mozilla Builders, which gave me the opportunity to further explore privacy-focused, local-first AI in the browser.
The interesting part is running LLMs directly in the browser. Instead of:
Browser → Server → AI API → Server → Browser
Apogee aims for:
Browser → Local Model → Response
This makes privacy the default rather than an afterthought.
I'm currently looking for contributors interested in browser extensions, WebGPU, local LLMs, UI/UX, performance, and documentation.
GitHub: https://github.com/darshi1337/apogee
If you're interested in local AI and privacy on the web, I'd love for you to check it out and contribute.
Top comments (1)
Firefox's Smart Window (which is built-in, not an addon) allows you to use a local LLM to power its features - including summarise etc.