Search credible GitHub repositories.

LLM training in simple, raw C/CUDA

Instant neural graphics primitives: lightning fast NeRF and more

Modern CUDA Learn Notes with PyTorch for Beginners, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.

A massively parallel, optimal functional runtime in Rust

DeepEP: an efficient expert-parallel communication library

DeepGEMM: clean and efficient BLAS kernel library on GPU

Code and data for paper "Deep Painterly Harmonization": https://arxiv.org/abs/1804.03189

The project is an official implementation of our CVPR2019 paper "Deep High-Resolution Representation Learning for Human Pose Estimation"

Fast parallel CTC.

No repository description provided.

[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.

Squeeze-and-Excitation Networks

Tile primitives for speedy kernels

how to optimize some algorithm in cuda.

This package contains the original 2012 AlexNet code.

No repository description provided.

CUDA Library Samples

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

cuGraph - RAPIDS Graph Analytics Library

Sample codes for my CUDA programming book

GPU Accelerated t-SNE for CUDA with Python bindings

NCCL Tests

Fully Convolutional Instance-aware Semantic Segmentation

No repository description provided.

Deformable ConvNets V2 (DCNv2) in PyTorch

[MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.

MatConvNet: CNNs for MATLAB

FSA/FST algorithms, differentiable, with PyTorch compatibility.

This is a series of GPU optimization topics. Here we will introduce how to optimize the CUDA kernel in detail. I will introduce several basic kernel optimizations, including: elementwise, reduce, sgemv, sgemm, etc. The performance of these kernels is basically at or near the theoretical limit.

RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.