NVIDIA / cub
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
See what the GitHub community is most excited about today.
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
cuGraph - RAPIDS Graph Analytics Library
cuVS - a library for vector search and clustering on the GPU
PyTorch bindings for CUTLASS grouped GEMM.
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.
Tile primitives for speedy kernels
Quantized Attention achieves speedup of 2-3x and 3-5x compared to FlashAttention and xformers, without lossing end-to-end metrics across language, image, and video models.
A massively parallel, optimal functional runtime in Rust
Lightning fast differentiable SSIM.
Causal depthwise conv1d in CUDA, with a PyTorch interface
NCCL Tests
CUDA accelerated rasterization of gaussian splatting