1.karpathy / llm.cLLM training in simple, raw C/CUDACudaMIT31K total3.8K forks+752 starsfrom 30,634 · score 6.62
2.xlite-dev / LeetCUDAAn Open-source book with Modern CUDA learning Notes for Beginners - BF16/FP8/FP4, HGEMM, CuTe, Flash-Attention, etc.cudahgemmcuda-kernelscuda-toolkitflash-attentioncuda-demo+6 moreCudaGPL-3.012K total1.3K forks+492 starsfrom 11,656 · score 6.20
3.NVlabs / instant-ngpInstant neural graphics primitives: lightning fast NeRF and moreneural-networkmachine-learningcudanerfcomputer-graphicscomputer-vision+6 moreCudaOther license18K total2.1K forks+102 starsfrom 17,484 · score 4.63
4.HigherOrderCO / HVM1HVM1 (2022): a massively parallel, optimal functional runtime in RustCudaMIT11K total437 forks+56 starsfrom 11,295 · score 4.04
5.rapidsai / cugraphcuGraph - RAPIDS Graph Analytics Libraryrapidsnvidiagpucudagraphgraph-algorithms+4 moreCudaAPACHE-2.02.3K total371 forks+46 starsfrom 2,210 · score 3.85
6.vlfeat / matconvnetMatConvNet: CNNs for MATLABCudaOther license1.4K total741 forks+2 starsfrom 1,425 · score 1.10