GPU Parallel Programming ML Frameworks

Tutorials, guides, and implementations for GPU computing using CUDA and related parallel processing frameworks. Focuses on learning CUDA fundamentals, optimization techniques, and GPU-accelerated computing. Does NOT include ML applications built with GPUs, collective communication libraries, or physics simulations—only the programming language/platform itself.

There are 61 gpu parallel programming frameworks tracked. 3 score above 70 (verified tier). The highest-rated is iree-org/iree at 73/100 with 3,655 stars. 6 of the top 10 are actively maintained.

Get all 61 projects as JSON

curl "https://pt-edge.onrender.com/api/v1/datasets/quality?domain=ml-frameworks&subcategory=gpu-parallel-programming&limit=20"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.

#	Framework	Score	Tier	Stars	Language
1	iree-org/iree A retargetable MLIR-based machine learning compiler and runtime toolkit.	73	Verified	3,655	C++
2	brucefan1983/GPUMD Graphics Processing Units Molecular Dynamics	73	Verified	738	Cuda
3	uxlfoundation/oneDAL oneAPI Data Analytics Library (oneDAL)	71	Verified	646	C++
4	rapidsai/cuml cuML - RAPIDS Machine Learning Library	69	Established	5,143	C++
5	NVIDIA/cutlass CUDA Templates and Python DSLs for High-Performance Linear Algebra	67	Established	9,426	C++
6	ROCm/Tensile [DEPRECATED] Moved to ROCm/rocm-libraries repo	64	Established	257	Python
7	NVIDIA/nccl Optimized primitives for collective multi-GPU communication	64	Established	4,521	C++
8	openucx/ucc Unified Collective Communication Library	61	Established	296	C
9	libxsmm/libxsmm Library for specialized dense and sparse matrix operations, and deep...	61	Established	944	C
10	ROCm/hipBLASLt [DEPRECATED] Moved to ROCm/rocm-libraries repo	60	Established	113	Assembly
11	uxlfoundation/oneCCL oneAPI Collective Communications Library (oneCCL)	60	Established	257	C++
12	PaddleJitLab/CUDATutorial A self-learning tutorail for CUDA High Performance Programing.	55	Established	911	JavaScript
13	google/gematria Machine learning for machine code.	53	Established	95	Python
14	mratsim/Arraymancer A fast, ergonomic and portable tensor library in Nim with a deep learning...	50	Established	1,391	Nim
15	XiaoMi/mace MACE is a deep learning inference framework optimized for mobile...	49	Emerging	5,033	C++
16	NVIDIA/GMAT A toolkit showing GPU's all-round capability in video processing	47	Emerging	194	C
17	cuMF/cumf_als CUDA Matrix Factorization Library with Alternating Least Square (ALS)	47	Emerging	181	Cuda
18	gorgonia/tensor package tensor provides efficient and generic n-dimensional arrays in Go...	46	Emerging	383	Go
19	srush/GPU-Puzzles Solve puzzles. Learn CUDA.	45	Emerging	11,992	Jupyter Notebook
20	mc2-project/mc2 A Platform for Secure Analytics and Machine Learning	44	Emerging	305	C++
21	MegEngine/MegCC MegCC是一个运行时超轻量，高效，移植简单的深度学习模型编译器	44	Emerging	484	C++
22	Edgecortix-Inc/mera A Heterogeneous Platform Deep Learning Compiler Framework from EdgeCortix	44	Emerging	35	Python
23	hshatti/Tensorium A platform agnostic fast tensor manipulation library using SIMD when...	43	Emerging	10	Pascal
24	OAID/TensorFlow-HRT Heterogeneous Run Time version of TensorFlow. Added heterogeneous...	42	Emerging	36	C++
25	bytedance/matxscript A high-performance, extensible Python AOT compiler.	42	Emerging	451	C++
26	MetaMachines/mm-ptx PTX Inject and Stack PTX	40	Emerging	2	C
27	OutofAi/cudacanvas Python Module for PyTorch Tensor Visualisation in CUDA Eliminating CPU Transfer	40	Emerging	41	Python
28	google/nccl-fastsocket NCCL Fast Socket is a transport layer plugin to improve NCCL collective...	40	Emerging	122	C++
29	eedalong/ECE408 Code base and slides for ECE408：Applied Parallel Programming On GPU.	39	Emerging	144	C++
30	Frikallo/axiom High-performance C++ tensor library with NumPy/PyTorch-like API	38	Emerging	102	C++
31	HenryNdubuaku/cuda-tutorials Comprehensive CUDA tutorials for Maths & ML with examples.	37	Emerging	215	Cuda
32	mratsim/laser The HPC toolbox: fused matrix multiplication, convolution, data-parallel...	37	Emerging	293	Nim
33	wangsiping97/FastGEMV High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.	36	Emerging	128	Cuda
34	AXERA-TECH/ax-npu-kit-650 AI algorithm SDK based on AX650	34	Emerging	6	C++
35	AndreSlavescu/meTile python-based eDSL for efficient Metal Shading Language code generation	34	Emerging	11	Python
36	openmlsys/openmlsys-cuda Tutorials for writing high-performance GPU operators in AI frameworks.	32	Emerging	134	Cuda
37	lawmurray/gpu-course Deep neural network and Adam optimizer in straight C and CUDA. Accompanies...	31	Emerging	4	C
38	Abrahamduru/mHC.cu 🚀 Implement mHC using CUDA for efficient Manifold-Constrained...	30	Emerging	1	Cuda
39	mikeroyal/OpenCL-Guide OpenCL Guide	30	Emerging	19	C++
40	priteshgohil/CUDA-programming-tutorial Get started with CUDA programming	29	Experimental	17	Cuda
41	SamerMakni/cuda-selector A simple tool to select the optimal CUDA device based custom criteria.	29	Experimental	8	Python
42	lcmialichi/php-cuda-ext Direct NVIDIA CUDA access for PHP. GPU-accelerated tensors, JIT-compiled...	28	Experimental	9	C
43	NumPower/numpower-autograd High performance PHP tensor with autograd (automatic differentiation) and...	28	Experimental	10	PHP
44	Venkat2811/yali Speed-of-Light SW efficiency by using ultra low-latency primitives for comms...	28	Experimental	13	Cuda
45	gabrielmaialva33/viva_tensor Pure Gleam tensor library with quantization (INT8, NF4, AWQ), Flash...	23	Experimental	2	Gleam
46	mrpottermusic/nccl-mesh-plugin 🌐 Enable distributed ML with the NCCL Mesh Plugin for efficient...	22	Experimental	—	C
47	realies/microgpt.c Karpathy's microgpt.py, in C	22	Experimental	—	C
48	porosh656/cuPDLPx 🚀 Accelerate your linear programming with cuPDLPx, a GPU-based solver that...	21	Experimental	—	Cuda
49	muhamadsafii-21/cutile-learn 🚀 Learn efficient CUDA programming with cuTile through hands-on tutorials...	21	Experimental	—	HTML
50	LessUp/cuda-kernel-academy CUDA Kernel Optimization Academy: SGEMM Tutorial, TensorCraft Ops, HPC...	21	Experimental	—	C++
51	LessUp/hpc-ai-optimization-lab CUDA HPC Kernel Optimization Textbook: Naive to Tensor Core — GEMM,...	21	Experimental	—	Cuda
52	nageshnnazare/cuda-know-hows cuda related stuff	20	Experimental	1	Cuda
53	Duconnor/Pudding This is the official repository for the project Pudding. Pudding enables you...	19	Experimental	4	C++
54	ProjectoOfficial/CUDA Learn cuda step-by-step starting from 0 with these simple and free code...	18	Experimental	2	Cuda
55	karton3c/kuda my custom open-source programing language	17	Experimental	1	Python
56	Pects1949/Cpp-Distributed-ML-Framework A C++ framework for distributed machine learning training, focusing on...	14	Experimental	—	CMake
57	aksayush2005/project-compiled A Mini Machine Learning Compiler with Hardware-Aware Optimization	14	Experimental	—	Python
58	garrettkinman/SteadyTensor An ultra-light, ultra-flexible tensor library written in pure Nim. Intended...	13	Experimental	—	Nim
59	rurumimic/cuda compute unified device architecture	13	Experimental	—	Cuda
60	camarababa/cuda-mastery-guide 🚀 Master CUDA programming with structured lessons covering fundamentals,...	13	Experimental	—	Cuda
61	takielias/php-tensor-for-windows PHP Tensor Extension for Windows https://github.com/RubixML/Tensor	13	Experimental	9	—

Comparisons in this category

oneDAL and oneCCL (71 vs 60)