Trending Projects

Run Claude AI agents on ultra-low-power chips with local-first memory and privacy.

Rust implementation of OpenClaw focusing on privacy-preserving AI model execution and security hardening.

Personal AI assistant platform with multi-app chat integration and extensible capabilities for local or cloud deployment.

Run frontier AI models locally across devices using RDMA and tensor parallelism

Open-source ChatGPT replacement with local AI model support

Cross-platform voice-to-text app with local & cloud AI models, privacy-first architecture

Lightweight AI assistant for ESP32 microcontrollers with GPIO, scheduling, custom tools, and memory.

Pure C inference engine for Mistral Voxtral 4B speech-to-text model with minimal dependencies

Fast local neural text-to-speech engine for offline voice synthesis

Reliable model swapping for local LLM servers - seamlessly switch between llama.cpp, vLLM, and compatible backends

Web UI for local AI with multiple backends and offline capabilities

Run local LLMs on any device with GPT4All

macOS offline speech-to-text app using local ML—no cloud, fully private voice dictation

Run LLMs locally with a single file

lib-pku/libpku: A collection of course materials and resources for vibe coders.

Run billion-parameter LLMs on embedded devices with extreme quantization for edge inference

Standalone Windows executables for Whisper speech-to-text & diarization without Python setup.