LLM training in simple, raw C/CUDA

DeepEP: an efficient expert-parallel communication library

DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling

CUDA accelerated rasterization of gaussian splatting

Tile primitives for speedy kernels

This package contains the original 2012 AlexNet code.

UNet diffusion model in pure CUDA

Reference implementation of Megalodon 7B model

Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity

Some CUDA example code with READMEs.