剪枝与蒸馏

🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

Training-free language model modification: Remove censorship, suppress slop, reduce sycophancy, and more

Awesome Knowledge Distillation

[CVPR 2023] DepGraph: Towards Any Structural Pruning; LLMs, Vision Foundation Models, etc.

SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime

Practical course about Large Language Models.

All-in-one training for vision models (YOLO, ViTs, RT-DETR, DINOv3): pretraining, fine-tuning, distillation.

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

TinyNeuralNetwork is an efficient and easy-to-use deep learning model compression framework.

Kandinsky 5.0: A family of diffusion models for Video & Image generation

Accelerate your Neural Architecture Search (NAS) through fast, reproducible and modular research.

A curated archive of breakthroughs in Agents, Architecture, Training, RAG, and On-Device AI.

Papers for deep neural network compression and acceleration

The Official Repo for "Quick Start Guide to Large Language Models"

模型压缩的小白入门教程，PDF下载地址 https://github.com/datawhalechina/awesome-compression/releases

Neural network model repository for highly sparse and sparse-quantized models with matching sparsification recipes

[NeurIPS 2024] SlimSAM: 0.1% Data Makes Segment Anything Slim

PMetal: high-performance Apple Silicon framework for local LLM inference, LoRA/QLoRA fine-tuning, serving, quantization, and MLX/Metal acceleration.

[ICLR 2025🔥] SVD-LLM & [NAACL 2025🔥] SVD-LLM V2

verl for a single consumer GPU. PPO, GRPO and on-policy distillation on NVIDIA GPUs.

A curated list of awesome edge machine learning resources, including research papers, inference engines, challenges, books, meetups and others.

[NeurIPS'26] Official Implementation for AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent(https://arxiv.org/abs/2602.03955)

FasterAI: Prune and Distill your models with FastAI and PyTorch

Enhancing LLMs with LoRA

📚 Collection of token-level model compression resources.