Trending
See what the GitHub community is most excited about today.
deepseek-ai / DeepEP
DeepEP: an efficient expert-parallel communication library
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.
nerfstudio-project / gsplat
CUDA accelerated rasterization of gaussian splatting
flashinfer-ai / flashinfer
FlashInfer: Kernel Library for LLM Serving
deepseek-ai / DeepGEMM
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
NVIDIA / cuopt
GPU accelerated decision optimization
BBuf / how-to-optim-algorithm-in-cuda
how to optimize some algorithm in cuda.
CannyLab / tsne-cuda
GPU Accelerated t-SNE for CUDA with Python bindings
HazyResearch / ThunderKittens
Tile primitives for speedy kernels
rapidsai / cuvs
cuVS - a library for vector search and clustering on the GPU
NVIDIA / cub
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
brucefan1983 / GPUMD
Graphics Processing Units Molecular Dynamics
NVIDIA / nccl-tests
NCCL Tests
NVlabs / instant-ngp
Instant neural graphics primitives: lightning fast NeRF and more
Dao-AILab / causal-conv1d
Causal depthwise conv1d in CUDA, with a PyTorch interface
rapidsai / 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.