LLM training in simple, raw C/CUDA

DeepEP: an efficient expert-parallel communication library

Sample codes for my CUDA programming book

[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl

Learn CUDA Programming, published by Packt

Examples demonstrating available options to program multiple GPUs in a single node or a cluster

CUDA Kernel Benchmarking Library

Graphics Processing Units Molecular Dynamics

Source code that accompanies The CUDA Handbook.

CUDA Learning guide

PopSift is an implementation of the SIFT algorithm in CUDA.

CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning

CUDA Data Parallel Primitives Library

EGGROLL in C, integer-first training

Alex Krizhevsky's original code from Google Code

FLAME GPU 2 is a GPU accelerated agent based modelling framework for CUDA C++ and Python

CUDA kernel author's tools

Parallel Simulated annealing in GPU using CUDA (used for floorplanning problem)

Header-only CUDA accelerated DNN library