Browser LLM inference on hand-written WGSL — up to a 35B sparse MoE, and models too big for one machine split by layer range across several over WebRTC. No TVM, no WebLLM runtime. Every number in BENCH.md.
AI in your browser that supports MCP Servers, no network call, no costs, 100% private
Free, private AI in your browser
High-performance, zero-dependency JS/TS Language Model & NLP suite for Node.js and browsers featuring text generation, self-attention heatmaps, RAG fact injection, and webpage autocomplete.
On-device RAG app (WebLLM + Transformers.js + pdf.js). Drop a PDF, ask questions, get cited answers. Runs fully in the browser; data stays local.
Scaffold for browser-local LLM chat apps. Runs Gemma 4 entirely in the browser via MediaPipe + WebGPU — no server, no cloud, no tokens. Mobile-first PWA, Atomic Design, i18n (DE/EN), multi-conversation. Fork it. Own it. Ship it.
A zero-dependency Node.js HTTPS-to-HTTP proxy that lets browser-based AI tools (Agile V Studio, Framewrk Studio) talk to local LLM backends like Ollama without mixed-content errors. Includes self-signed TLS, CORS, token tracking dashboard, Docker support, and OpenAI/Anthropic-compatible passthrough.
Run small LLMs directly in your web browser, no cloud computing needed.
Terminal-style portfolio website that runs a local LLM in your browser. Type `ask` to chat with an on-device LLM (Gemini Nano / WebLLM via WebGPU) — no backend, no API keys, your messages stay local. Built with React + Vite.
Starling - Lightning Fast Autofill Extension
Browser-native LLM chat powered by Transformers.js, WebAssembly and WebGPU.
Offline-first AI assistant (browser LLM + RAG) for newcomers & disconnected regions. No API keys, works in airplanemode.
Pushing fused WebGPU transformer kernels to max model size — int4, tiled FFN, Phi-3-mini 3.6B in Chrome
在浏览器中用 WebGPU 全本地运行 DeepSeek-R1-Distill-Qwen-1.5B 推理模型，无需服务器、数据不出浏览器。基于 Transformers.js + ONNX Runtime Web，支持流式输出、思考过程可视化、Markdown 与数学公式渲染，代码含详尽中文注释，适合学习 WebGPU / Web Worker 工程实践。
Privacy-first document editor with local in-browser LLM inference via HuggingFace Transformers.js, WebGPU, and Tiptap.
To associate your repository with the browser-llm topic, visit your repo's landing page and select "manage topics."