offline-ai

Run Claude Code 100% on-device with local AI on Apple Silicon. MLX-native Anthropic-API server. 6 fighters incl. Muse-Glimmer 30B (now multimodal — reads images, abliterated), Gemma 4 31B, Qwen 3.5 122B (65 tok/s), DeepSeek V4 Flash (1M ctx). Private, offline, airgap-ready. Built for NDA / legal / healthcare workflows.

The Swiss Army Knife of Offline AI. Chat, see, speak, and generate images on your phone or Mac — GGUF LLMs, vision, Whisper speech-to-text, Stable Diffusion, tool calling, and local-network servers. Runs on your CPU, GPU, or NPU. No account, no API key, zero data leaves your device.

U-Claw 虾盘 — OpenClaw AI 助手离线安装 U 盘：一键装好 Claude Code/Codex/OpenClaw，无需联网即插即用；新品 U-King AI 装机管家详见 u-king.org，另有远程支持与定制 AI 开发。

NativeMind: Your fully private, open-source, on-device AI assistant

HiKid - Your AI English Pal. A desktop app built with React and TypeScript, targeting children in non-English-speaking countries who want to practice English speaking and listening.

Android 17 local LLM prototype with Jetpack Compose and ONNX Runtime for offline AI inference experiments.

Open-source AI browser agent for Chrome and Firefox (monorepo) 🧠

Open-source local-first AI agent for desktop work. No account, no telemetry: use local models with Ollama/Rapid-MLX or bring your own provider key.

LeanType is an Open-source keyboard, fork of HeliBoard with additional features and customisation

refugiOS es un sistema operativo portátil diseñado para situaciones de emergencia, falta de conectividad a Internet o necesidad extrema de privacidad

Local-first AI-powered document intelligence platform for investigative journalism

React Native Apple LLM plugin using Foundation Models

Open-source local AI workspace — advancing on-device inference.

破限本地AI女友后宫，openclaw/claude code+画图语音向量数据库+live2D+桌宠+酒馆角色卡导入+前端，QQ+Telegram双通道，8G显存可跑🩵uncensored Fully offline AI girlfriends harem Openclaw/Claude code+Local LLM+GPT-SoVITS+ComfyUI image+Live2D+desktop pet+SilllyTavern Character card import+frontend | Dual channels for QQ & Telegram | Dynamic 8G VRAM scheduling+mem0 qdrant, can run offline

One-click Qwen3.6-27B inference on Windows. 158 tok/s on RTX 5090, 72 tok/s on RTX 3090. Native, no WSL, no Docker, no telemetry.

Offline crisis companion for Bangladesh floods. Signed phone-to-phone SOS, flood-avoiding shelter routing, first aid and emergency translation for six languages Google Translate doesn't carry — no internet, no server, no account. Gemma 4 runs on the handset.

True on-device AI for Kotlin Multiplatform (Android, iOS, Desktop, JVM, WASM). LLM, Speech-to-Text and Image Generation — powered by llama.cpp, whisper.cpp and stable-diffusion.cpp.

[Deprecated] Zero-configuration local AI studio. Active development has moved to Uncensored-Local-Studio.

Free offline AI assistant for Windows — reads your screen, hears your question, speaks the answer. No API key, runs locally with Ollama.

Run a <400ms latency Voice Agent on just 4GB VRAM. Fully offline, no API keys required. Optimized for GTX 1650 and edge robotics with zero-copy inference. (Apache 2.0)