Git Stars Git Stars

Cuda Repositories

Explore 65 active open source Cuda repositories on GitHub with a combined 165,484 stars (averaging 2,545 stars per project). The #1 ranked Cuda project is karpathy/llm.c with 30,323 stars.

65
Total Repositories
165,484
Total Stars
2,545
Avg Stars/Repo
1
karpathy/llm.c

LLM training in simple, raw C/CUDA

karpathy
โญ 30,323 Cuda ๐Ÿ‘ค karpathy User ๐Ÿ• Updated 2025-06-26 ๐Ÿ“… Created 2024-04-08
2
NVlabs/instant-ngp

Instant neural graphics primitives: lightning fast NeRF and more

NVlabs
โญ 17,606 Cuda ๐Ÿ‘ค NVlabs Organization ๐Ÿ• Updated 2026-02-02 ๐Ÿ“… Created 2022-01-05
3
HigherOrderCO/HVM1

HVM1 (2022): a massively parallel, optimal functional runtime in Rust

HigherOrderCO
โญ 11,375 Cuda ๐Ÿ‘ค HigherOrderCO Organization ๐Ÿ• Updated 2026-09-16 ๐Ÿ“… Created 2022-01-03
4
deepseek-ai/DeepEP

DeepEP: an efficient expert-parallel communication library

deepseek-ai
โญ 10,462 Cuda ๐Ÿ‘ค deepseek-ai Organization ๐Ÿ• Updated 2026-11-20 ๐Ÿ“… Created 2025-02-17
5
deepseek-ai/DeepGEMM

DeepGEMM: clean and efficient BLAS kernel library on GPU

deepseek-ai
โญ 9,210 Cuda ๐Ÿ‘ค deepseek-ai Organization ๐Ÿ• Updated 2026-11-19 ๐Ÿ“… Created 2025-02-13
6
flashinfer-ai/flashinfer

FlashInfer: Kernel Library for LLM Serving

flashinfer-ai
โญ 6,982 Cuda ๐Ÿ‘ค flashinfer-ai Organization ๐Ÿ• Updated 2026-11-21 ๐Ÿ“… Created 2023-07-22
7
leoxiaobin/deep-high-resolution-net.pytorch

The project is an official implementation of our CVPR2019 paper "Deep High-Resolution Representation Learning for Human Pose Estimation"

leoxiaobin
โญ 4,457 Cuda ๐Ÿ‘ค leoxiaobin User ๐Ÿ• Updated 2024-08-30 ๐Ÿ“… Created 2019-02-25
9
baidu-research/warp-ctc

Fast parallel CTC.

baidu-research
โญ 4,079 Cuda ๐Ÿ‘ค baidu-research Organization ๐Ÿ• Updated 2024-03-04 ๐Ÿ“… Created 2016-01-14
10
thu-ml/SageAttention

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

thu-ml
โญ 4,017 Cuda ๐Ÿ‘ค thu-ml Organization ๐Ÿ• Updated 2026-01-17 ๐Ÿ“… Created 2024-10-03
11
HazyResearch/ThunderKittens

Tile primitives for speedy kernels

HazyResearch
โญ 3,805 Cuda ๐Ÿ‘ค HazyResearch Organization ๐Ÿ• Updated 2026-09-12 ๐Ÿ“… Created 2024-03-04
12
BBuf/how-to-optim-algorithm-in-cuda

how to optimize some algorithm in cuda.

BBuf
โญ 3,356 Cuda ๐Ÿ‘ค BBuf User ๐Ÿ• Updated 2026-11-02 ๐Ÿ“… Created 2018-07-18
13
computerhistory/AlexNet-Source-Code

This package contains the original 2012 AlexNet code.

computerhistory
โญ 2,882 Cuda ๐Ÿ‘ค computerhistory Organization ๐Ÿ• Updated 2025-03-12 ๐Ÿ“… Created 2024-12-09
15
mirage-project/mirage

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

mirage-project
โญ 2,625 Cuda ๐Ÿ‘ค mirage-project Organization ๐Ÿ• Updated 2026-11-21 ๐Ÿ“… Created 2024-05-08
16
NVIDIA/CUDALibrarySamples

CUDA Library Samples

NVIDIA
โญ 2,555 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-21 ๐Ÿ“… Created 2019-05-09
17
rapidsai/cugraph

cuGraph - RAPIDS Graph Analytics Library

rapidsai
โญ 2,260 Cuda ๐Ÿ‘ค rapidsai Organization ๐Ÿ• Updated 2026-11-20 ๐Ÿ“… Created 2018-11-15
18
brucefan1983/CUDA-Programming

Sample codes for my CUDA programming book

brucefan1983
โญ 2,106 Cuda ๐Ÿ‘ค brucefan1983 User ๐Ÿ• Updated 2025-12-14 ๐Ÿ“… Created 2019-05-03
19
CannyLab/tsne-cuda

GPU Accelerated t-SNE for CUDA with Python bindings

CannyLab
โญ 1,955 Cuda ๐Ÿ‘ค CannyLab Organization ๐Ÿ• Updated 2026-07-22 ๐Ÿ“… Created 2018-03-24
โญ 1,711 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-10-05 ๐Ÿ“… Created 2017-08-08
21
mit-han-lab/torchsparse

[MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.

mit-han-lab
โญ 1,460 Cuda ๐Ÿ‘ค mit-han-lab Organization ๐Ÿ• Updated 2025-02-24 ๐Ÿ“… Created 2020-08-24
22
k2-fsa/k2

FSA/FST algorithms, differentiable, with PyTorch compatibility.

k2-fsa
โญ 1,371 Cuda ๐Ÿ‘ค k2-fsa Organization ๐Ÿ• Updated 2026-11-13 ๐Ÿ“… Created 2020-04-17
โญ 1,362 Cuda ๐Ÿ‘ค graphdeco-inria Organization ๐Ÿ• Updated 2024-10-21 ๐Ÿ“… Created 2023-07-04
24
MoonshotAI/FlashKDA

FlashKDA: high-performance Kimi Delta Attention kernels

MoonshotAI
โญ 1,300 Cuda ๐Ÿ‘ค MoonshotAI Organization ๐Ÿ• Updated 2026-11-16 ๐Ÿ“… Created 2026-04-20
25
siboehm/SGEMM_CUDA

Fast CUDA matrix multiplication from scratch

siboehm
โญ 1,297 Cuda ๐Ÿ‘ค siboehm User ๐Ÿ• Updated 2025-09-02 ๐Ÿ“… Created 2022-11-13
26
thu-ml/SpargeAttn

[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.

thu-ml
โญ 1,269 Cuda ๐Ÿ‘ค thu-ml Organization ๐Ÿ• Updated 2026-02-25 ๐Ÿ“… Created 2025-02-25
27
unum-cloud/UForm

Distributed training & inference for multimodal perception: shared embeddings across text, images, video, audio, code & graphs from 64 to 1024 dimensions ๐Ÿ–ผ๏ธ

unum-cloud
โญ 1,253 Cuda ๐Ÿ‘ค unum-cloud Organization ๐Ÿ• Updated 2026-11-18 ๐Ÿ“… Created 2023-02-21
28
NVIDIA/cuopt

GPU accelerated decision optimization

NVIDIA
โญ 1,123 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-20 ๐Ÿ“… Created 2025-04-08
29
Tongkaio/CUDA_Kernel_Samples

CUDA Operators: Manual Implementation and Interview Guide

Tongkaio
โญ 1,091 Cuda ๐Ÿ‘ค Tongkaio User ๐Ÿ• Updated 2025-08-23 ๐Ÿ“… Created 2024-04-30
30
NVIDIA/raft

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.

NVIDIA
โญ 1,055 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-20 ๐Ÿ“… Created 2019-06-13
31
tspeterkim/flash-attention-minimal

Flash Attention in ~100 lines of CUDA (forward pass only)

tspeterkim
โญ 1,036 Cuda ๐Ÿ‘ค tspeterkim User ๐Ÿ• Updated 2024-12-30 ๐Ÿ“… Created 2024-03-07
32
nv-tlabs/NKSR

[CVPR 2023 Highlight] Neural Kernel Surface Reconstruction

nv-tlabs
โญ 1,012 Cuda ๐Ÿ‘ค nv-tlabs Organization ๐Ÿ• Updated 2026-08-25 ๐Ÿ“… Created 2023-06-08
33
NVIDIA/cudnn-frontend

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

NVIDIA
โญ 1,008 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-21 ๐Ÿ“… Created 2021-01-22
34
Celebrandil/CudaSift

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

Celebrandil
โญ 962 Cuda ๐Ÿ‘ค Celebrandil User ๐Ÿ• Updated 2026-03-17 ๐Ÿ“… Created 2013-06-20
35
NVIDIA/nvbench

CUDA Kernel Benchmarking Library

NVIDIA
โญ 952 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-21 ๐Ÿ“… Created 2021-03-03
36
olcf/cuda-training-series

Training materials associated with NVIDIA's CUDA Training Series (www.olcf.ornl.gov/cuda-training-series/)

olcf
โญ 927 Cuda ๐Ÿ‘ค olcf Organization ๐Ÿ• Updated 2024-08-19 ๐Ÿ“… Created 2020-01-06
37
NVIDIA/multi-gpu-programming-models

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

NVIDIA
โญ 919 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-06 ๐Ÿ“… Created 2018-05-11
38
NVIDIA/cuvs

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

NVIDIA
โญ 902 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-21 ๐Ÿ“… Created 2023-11-20
39
brucefan1983/GPUMD

Graphics Processing Units Molecular Dynamics

brucefan1983
โญ 868 Cuda ๐Ÿ‘ค brucefan1983 User ๐Ÿ• Updated 2026-11-21 ๐Ÿ“… Created 2017-07-14
โญ 836 Cuda ๐Ÿ‘ค princeton-vl Organization ๐Ÿ• Updated 2025-05-10 ๐Ÿ“… Created 2021-03-23
41
jinbooooom/ai-infra-hpc

HPC tutorial, covering collective communication (MPI, NCCL), CUDA programming, vectorization SIMD, RDMA communication, etc.

jinbooooom
โญ 795 Cuda ๐Ÿ‘ค jinbooooom User ๐Ÿ• Updated 2026-04-27 ๐Ÿ“… Created 2024-07-23
42
19reborn/NeuS2

[ICCV 2023] Official code for NeuS2

19reborn
โญ 713 Cuda ๐Ÿ‘ค 19reborn User ๐Ÿ• Updated 2024-03-22 ๐Ÿ“… Created 2022-12-07
43
NVIDIA/AMGX

Distributed multigrid linear solver library on GPU

NVIDIA
โญ 707 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-07-09 ๐Ÿ“… Created 2017-10-04
โญ 679 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-21 ๐Ÿ“… Created 2019-12-09
45
clu0/unet.cu

UNet diffusion model in pure CUDA

clu0
โญ 658 Cuda ๐Ÿ‘ค clu0 User ๐Ÿ• Updated 2024-06-28 ๐Ÿ“… Created 2024-06-28
46
theialab/radfoam

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

theialab
โญ 657 Cuda ๐Ÿ‘ค theialab Organization ๐Ÿ• Updated 2025-05-14 ๐Ÿ“… Created 2025-02-04
47
tensorflow/recommenders-addons

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

tensorflow
โญ 635 Cuda ๐Ÿ‘ค tensorflow Organization ๐Ÿ• Updated 2025-09-04 ๐Ÿ“… Created 2020-11-20
48
pranjalssh/fast.cu

Fastest kernels written from scratch

pranjalssh
โญ 630 Cuda ๐Ÿ‘ค pranjalssh User ๐Ÿ• Updated 2026-08-15 ๐Ÿ“… Created 2024-11-04
49
DavidDiazGuerra/gpuRIR

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

DavidDiazGuerra
โญ 617 Cuda ๐Ÿ‘ค DavidDiazGuerra User ๐Ÿ• Updated 2026-08-25 ๐Ÿ“… Created 2018-10-08
50
NVIDIA/nvshmem

NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process communication and coordination overheads by allowing programmers to perform one-sided communication from within CUDA kernels and on CUDA streams.

NVIDIA
โญ 606 Cuda ๐Ÿ‘ค NVIDIA Organization ๐Ÿ• Updated 2026-11-12 ๐Ÿ“… Created 2025-08-27

Frequently Asked Questions

1

What is the most starred Cuda repository on GitHub?

The most starred active Cuda repository on GitHub is karpathy/llm.c with 30,323 stars. LLM training in simple, raw C/CUDA

2

How many popular Cuda projects are tracked on Git Stars?

We currently track 65 active open source Cuda repositories with 500+ stars, accumulating 165,484 total stars and an average of 2,545 stars per project.

3

How often is the Cuda GitHub ranking updated?

All Cuda repository star counts, activity dates, and rankings are synchronized daily from the official GitHub API.