Cuda

RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.

GPU accelerated decision optimization

[CVPR 2023 Highlight] Neural Kernel Surface Reconstruction

A CUDA implementation of SIFT for NVidia GPUs (1.2 ms on a GTX 1060)

CUDA Kernel Benchmarking Library

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

USDT钱包靓号生成器,Trx靓号,波场GPU靓号源码,TRON钱包靓号生成,波场地址靓号生成器

cuVS - a library for vector search and clustering on the GPU

Graphics Processing Units Molecular Dynamics

[ICCV 2023] Official code for NeuS2

hpc 教程，包含集合通信(mpi、nccl)、cuda 编程、向量化 SIMD、RDMA 通信等

Distributed multigrid linear solver library on GPU

Original implementation of "Radiant Foam: Real-Time Differentiable Ray Tracing"

UNet diffusion model in pure CUDA

Additional utils and helpers to extend TensorFlow when build recommendation systems, contributed and maintained by SIG Recommenders.

100 days of building GPU kernels!

Fastest kernels written from scratch

Published on Nature Machine Intelligence! The first real robot(quadrotor) based on differentiable physics training.

Python library for Room Impulse Response (RIR) simulation with GPU acceleration

Source code that accompanies The CUDA Handbook.

Several optimization methods of half-precision general matrix multiplication (HGEMM) using tensor core with WMMA API and MMA PTX instruction.

CUDA Learning guide

Distribution-Aware Coordinate Representation for Human Pose Estimation

Static suckless single batch CUDA-only qwen3-0.6B mini inference engine

flash attention tutorial written in python, triton, cuda, cutlass

Reference implementation of Megalodon 7B model

A tutorial for CUDA&PyTorch