TTS with kokoro and onnx runtime
Python client for the Telegram's tdlib
Phi-3.5 for Mac: Locally-run Vision and Language Models
VoiceStudio is the open-source, fully-local ElevenLabs alternative
Apple Silicon (MLX) port of Karpathy's autoresearch
State-of-the-art TTS model under 25MB
LLM inference server with continuous batching & SSD caching
Offline knowledge library for survival and rebuilding
A natural language interface for computers
Simulation framework for accelerating research
An open phone agent model & framework
Run your own AI cluster at home with everyday devices
A GPT-4o Level MLLM for Vision, Speech and Multimodal Live Streaming
Run LLMs with MLX
Low-latency AI inference engine optimized for mobile devices
General proxy performance testing tool based on Clash using Telegram
A text-to-speech, speech-to-text and speech-to-speech library
A Telegram bot that integrates with OpenAI's official ChatGPT APIs
Let your agent control your phone
Stable Diffusion web UI
Native MLX runtime for Laya typed decision models
Find the local LLM that actually runs and performs best
Datasets, transforms and models specific to Computer Vision
MiniCPM5: SOTA on-device LLMs, small yet powerful
Reading book source