Fast and memory-efficient exact attention

PArallel Distributed Deep LEarning （『飞桨』核心框架，高性能单机、分布式训练和跨平台部署）

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper GPUs, to provide better performance with lower memory utilization in bot…

CUDA Library Samples

This repository consists for gpu bootcamp material for HPC and AI

Deep Scalable Sparse Tensor Network Engine (DSSTNE) is an Amazon developed library for building Deep Learning (DL) machine learning (ML) models

TensorRT is a C++ library for high performance inference on NVIDIA GPUs and deep learning accelerators.

Documentations for PaddlePaddle

An implementation of a deep learning recommendation model (DLRM)

CUDA Templates for Linear Algebra Subroutines

NumPy-like API accelerated with CUDA