Industry leading face manipulation platform
Run Local LLMs on Any Device. Open-source
Personal AI, On Personal Devices
AirLLM 70B inference with single 4GB GPU
FlashInfer: Kernel Library for LLM Serving
Machine learning on FPGAs using HLS
Fast stable diffusion on CPU and AI PC
VoiceStudio is the open-source, fully-local ElevenLabs alternative
Universal LLM Deployment Engine with ML Compilation
Run Bonsai (1-bit) and Ternary-Bonsai language models locally
Python framework for building scalable multi-agent systems
Official repository for LTX-Video
TTS with kokoro and onnx runtime
DeepMind's software stack for physics-based simulation
Understanding of AI Infra: Quantitative Analysis and System Design
AI video generator optimized for low VRAM and older GPUs use
Apple Silicon (MLX) port of Karpathy's autoresearch
The most powerful local music generation model
Secure local-first microVM sandbox for running untrusted code fast
Official inference framework for 1-bit LLMs
High-Resolution Image Synthesis with Latent Diffusion Models
A Domain-Fronting Relay that routes traffic though GAS
Any model. Any hardware. Zero compromise
Parallax is a distributed model serving framework
High-performance Inference and Deployment Toolkit for LLMs and VLMs