deepseek-ai / DeepGEMM
DeepGEMM: clean and efficient BLAS kernel library on GPU
See what the GitHub community is most excited about today.
DeepGEMM: clean and efficient BLAS kernel library on GPU
RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.
[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.
how to optimize some algorithm in cuda.
DeepEP: an efficient expert-parallel communication library
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
Causal depthwise conv1d in CUDA, with a PyTorch interface
Mirage Persistent Kernel: Compiling LLMs into a MegaKernel
A plugin to use Nvidia GPU in PySCF package
cuVS - a library for vector search and clustering on the GPU
SplatAD: Real-Time Lidar and Camera Rendering with 3D Gaussian Splatting for Autonomous Driving
NCCL Tests
Tile primitives for speedy kernels