Trending
See what the GitHub community is most excited about this week.
Spoken Language: Any
Language: Cuda
Date range: This week
deepseek-ai / DeepEP
DeepEP: an efficient expert-parallel communication library
NVlabs / instant-ngp
Instant neural graphics primitives: lightning fast NeRF and more
JeffreyXiang / CuMesh
Cuda mesh utils.
NVIDIA / nccl-tests
NCCL Tests
Dao-AILab / causal-conv1d
Causal depthwise conv1d in CUDA, with a PyTorch interface
HazyResearch / ThunderKittens
Tile primitives for speedy kernels
BBuf / how-to-optim-algorithm-in-cuda
how to optimize some algorithm in cuda.
alibaba / rtp-llm
RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.
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.
NVIDIA / cub
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
siboehm / SGEMM_CUDA
Fast CUDA matrix multiplication from scratch