Artifacts of EVT ASPLOS'24

FlagGems is an operator library for large language models implemented in Triton Language.

Memory-efficient multi layer perceptron implementation in OpenAI Triton.

Helpful tools and examples for working with flex-attention

[NeurIPS'24 Spotlight] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention, which reduces inference latency by up to 10x for pre-filling on an A100 whil…

A Subscribe Convert Tool for Clash

Samples for CUDA Developers which demonstrates features in CUDA Toolkit

C++ extensions in PyTorch

how to optimize some algorithm in cuda.

compiler learning resources collect.

real Transformer TeraFLOPS on various GPUs

Development repository for the Triton language and compiler

A local chatbot fine-tuned by bilibili user comments.

A self-learning tutorail for CUDA High Performance Programing.

高性能计算相关知识学习笔记，包含学习笔记和相关知识的代码demo，在持续完善中。 如果有帮助的话请Star一下，对作者帮助很大，谢谢！

Compiler Infrastructure for Neural Networks

A Easy-to-understand TensorOp Matmul Tutorial

Practice on cifar100(ResNet, DenseNet, VGG, GoogleNet, InceptionV3, InceptionV4, Inception-ResNetv2, Xception, Resnet In Resnet, ResNext,ShuffleNet, ShuffleNetv2, MobileNet, MobileNetv2, SqueezeNet…

My solutions to the Glomers Challenge: a series of distributed systems challenges.

Materials for the Learn PyTorch for Deep Learning: Zero to Mastery course.

A Zsh theme

A high-throughput and memory-efficient inference and serving engine for LLMs

Abstraction your words——never mind the scandal and liber

一个纯前端的抽象话转换器

Fast and memory-efficient exact attention

Fast Synchronization-Free Algorithms for Parallel Sparse Triangular Solves with Multiple Right-Hand Sides (SpTRSM)

AISystem 主要是指AI系统，包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术

Transformer related optimization, including BERT, GPT