A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

A fast GPU memory copy library based on NVIDIA GPUDirect RDMA technology

FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/

Material for gpu-mode lectures

LLM inference in C/C++

The Art of Debugging

📰 Must-read papers on KV Cache Compression (constantly updating 🤗).

Magnificent app which corrects your previous console command.

🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!

A faster int-to-int hashmap implemented in C++.

Concurrency primitives, safe memory reclamation mechanisms and non-blocking (including lock-free) data structures designed to aid in the research, design and implementation of high performance conc…

LLM training in simple, raw C/CUDA

Systems for GenAI

Development repository for the Triton language and compiler

Boosting 4-bit inference kernels with 2:4 Sparsity

A basic deep learning library, comparable to a very minimal version of PyTorch.

µcoro

QRec: A Python Framework for quick implementation of recommender systems (TensorFlow Based)

练习下用pytorch来复现下经典的推荐系统模型, 如MF, FM, DeepConn, MMOE, PLE, DeepFM, NFM, DCN, AFM, AutoInt, ONN, FiBiNET, DCN-v2, AFN, DCAP等

Notes about courses Dive into Deep Learning by Mu Li

A model compilation solution for various hardware

row-major matmul optimization

how to optimize some algorithm in cuda.

C++ library for executors

AISystem 主要是指AI系统，包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术

关于Transformer模型的最简洁pytorch实现，包含详细注释

🧡 Follow everything in one place

LMDeploy is a toolkit for compressing, deploying, and serving LLMs.

Pytorch domain library for recommendation systems