Trending
See what the GitHub community is most excited about today.
Spoken Language: Any
Language: Cuda
Date range: Today
NVIDIA / nccl-tests
NCCL Tests
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 / cuopt
GPU accelerated decision optimization
HazyResearch / ThunderKittens
Tile primitives for speedy kernels
NVIDIA / cub
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
NVIDIA / nvbench
CUDA Kernel Benchmarking Library
Dao-AILab / causal-conv1d
Causal depthwise conv1d in CUDA, with a PyTorch interface
deepseek-ai / DeepEP
DeepEP: an efficient expert-parallel communication library
karpathy / llm.c
LLM training in simple, raw C/CUDA
alibaba / rtp-llm
RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.
NVlabs / instant-ngp
Instant neural graphics primitives: lightning fast NeRF and more
deepseek-ai / DeepGEMM
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
JeffreyXiang / CuMesh
Cuda mesh utils.