Offline chat on desktop or Android via Termux with Prefrontal
What they built
Prefrontal is an open-source chat interface for local models. When run with Ollama or llama.cpp, no chat data leaves your machine or LAN. OpenRouter is an optional cloud mode.
Setup
- Node.js app at http://localhost:3000
- Runtime: Ollama (desktop) or llama.cpp (any platform, including Android/Termux)
- Suggested models: gemma3:4b (recommended), gemma3:12b, llama3.2, mistral
Gotchas
- The hosted live demo only supports OpenRouter and stores your key in a cookie; it's for previewing the UI, not private use
Do it yourself
- ollama serve & then ollama pull gemma3:4b
- git clone the repo, npm install && npm start
- Open http://localhost:3000 → Settings → Runtime: Ollama, URL http://localhost:11434, model gemma3:4b
git clone https://github.com/sidx1-scratch/prefrontal
cd prefrontal
npm install && npm startRelated
A private ChatGPT for a family of four on a Mac mini
One always-on M4 Mac mini serves Open WebUI to every phone and laptop in the house. Kids get a tutor preset, and three AI subscriptions are gone.
Øxygen: a private desktop chat app for Ollama
Conversations, files and settings live in a file in your home directory. It talks only to Ollama on 127.0.0.1 and adds optional embeddings, vision, voice and image generation.
Branching conversation graphs with OpenYoke
A Tauri and Rust desktop app for Ollama where every exchange is a node you can branch from, and sibling branches never bleed context into each other.
Offline LLM on a Raspberry Pi 5 for classification and extraction
Phi-3 mini at Q4_K_M runs at 4-6 tokens per second on an 8 GB Pi 5 with no network at all. Under GDPR, data that never leaves the board isn't transferred anywhere.