br-AI-n

A fully private, offline AI assistant running local LLMs in the browser using WebAssembly and WebGPU.

The ultimate privacy-first AI companion built directly into your Chrome Side Panel. br-AI-n runs Large Language Models (LLMs) privately and natively in your browser. Powered by cutting-edge WebAssembly (WASM) and WebGPU hardware acceleration, your AI assistant runs directly on your device — no API keys, no monthly fees, and zero hidden tracking. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 🛡️ 100% PRIVATE & LOCAL-FIRST ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Your conversations, documents, and code never leave your machine: ▸ Local Browser Models — Chats and prompts are processed strictly inside your browser's sandboxed environment. ▸ Local Server Bridge — Connect directly to your own self-hosted endpoints with complete confidentiality. ▸ Fail-Closed Security — Network tools are protected by a strict kill-switch that blocks external requests completely while Web Tools are off. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ⚡ HIGH-PERFORMANCE NATIVE EXECUTION ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ▸ WebGPU Acceleration — Taps into your GPU for smooth, responsive local inference. ▸ Smart CPU Fallback — Optimized single-threaded WebAssembly (SIMD) ensures broad hardware compatibility. ▸ Instant Offline Caching — Models download once into your local browser cache for instant startup anytime. ▸ Transparent Open Weights — Direct attribution and license transparency for all supported open models. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 🔌 BRING YOUR OWN LOCAL SERVER ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Already have a local AI workstation? br-AI-n connects to local server backends (including Ollama, LM Studio, and llama.cpp). Offload intensive workloads to your dedicated rig while enjoying a seamless Chrome interface. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 🎭 CUSTOMIZABLE PERSONAS ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Tailor your AI assistant to any task in seconds: ▸ Built-in Presets — Assistant, Code Ninja, Summarizer, Translator, and Grammar Checker. ▸ Custom Personas — Create and save unlimited personas with custom system prompts, response styles, and formatting rules. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 📝 FLOATING INLINE QUICK-ACTION WIDGET ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Highlight text on any webpage for instant inline AI actions: ▸ 1-Click Transformations — Summarize, explain complex concepts, or fix grammar without leaving the page. ▸ Context-Aware Chat — Ask follow-up questions directly inside the floating card. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 🧠 OPTIONAL WEB TOOLS (OFF BY DEFAULT) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Turn your local model into an autonomous research agent when you need live facts: ▸ Smart Web Search — Accesses user-configured search engines, public encyclopedias, open academic archives, and live global news feeds. ▸ Tab Reader — Read the contents of your active tab on command (the extension never crawls websites autonomously). ▸ Sandboxed Code Execution — Safely runs calculations and code snippets in an isolated runtime. *Web Tools are 100% optional and disabled by default to guarantee full offline privacy.* ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 📄 SEAMLESS ACTIVE TAB CONTEXT ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Reading a long article or documentation? Click "Attach Page Context" to pull the page text directly into your chat. Instantly summarize long-form articles, extract key points, or ask specific questions about the open page. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ✨ DESIGNED FOR WORKFLOWS ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ▸ Native Reasoning Display — Elegantly renders expandable thought steps from advanced reasoning models. ▸ Persistent Side Panel — Always accessible with a single click, perfectly integrated into your browsing flow. Take back control of your data. Experience decentralized, private browser AI with br-AI-n.