tvm-quantization

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Universal LLM Deployment Engine with ML Compilation

中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs)

Open deep learning compiler stack for cpu, gpu and specialized accelerators

Lossy PNG compressor — pngquant command based on libimagequant library

An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

Fast inference engine for Transformer models

Bringing stable diffusion models to web browsers. Everything runs inside the browser with no server support.

Vector (and Scalar) Quantization, in Pytorch

Sparsity-aware deep learning inference runtime for CPUs