Popular GitHub Repositories for Compression

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

AIMET is a library that provides advanced quantization and compression techniques for trained neural network models.

A curated list for Efficient Large Language Models

PaddleSlim is an open-source library for deep model compression and architecture search.

A toolkit to optimize ML models for deployment for Keras and TensorFlow, including quantization and pruning.

A PyTorch library and evaluation platform for end-to-end compression research

Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM

[CVPR 2020] GAN Compression: Efficient Architectures for Interactive Conditional GANs

PyTorch Implementation of "Lossless Image Compression through Super-Resolution"

[NeurIPS 2023] LLM-Pruner: On the Structural Pruning of Large Language Models. Support Llama-3/3.1, Llama-2, LLaMA, BLOOM, Vicuna, Baichuan, TinyLlama, etc.

This repository collects papers for "A Survey on Knowledge Distillation of Large Language Models". We break down KD into Knowledge Elicitation and Distillation Algorithms, and explore the Skill & Vertical Distillation of LLMs.

collection of works aiming at reducing model sizes or the ASIC/FPGA accelerator for machine learning

A collection of tools for neural compression enthusiasts.

[NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization

A Compressed Stable Diffusion for Efficient Text-to-Image Generation [ECCV'24]

[CVPR'20] ZeroQ: A Novel Zero Shot Quantization Framework

模型压缩的小白入门教程

LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networks

NN based lossless compression