A Quirky Assortment of CuTe Kernels

Distributed Compiler based on Triton for Parallel Systems

🚀 Efficient implementations of state-of-the-art linear attention models

PyTorch Single Controller

Fastest CRC32 for x86, Intel and AMD, + comprehensive derivation and discussion of various approaches

Bindings for RDMA ibverbs through rdma-core

A high-performance distributed file system designed to address the challenges of AI training and inference workloads.

AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.

Foundation Architecture for (M)LLMs

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…

Hackable and optimized Transformers building blocks, supporting a composable construction.

Multithreaded Python without the GIL

Text describing xv6 on RISC-V

Transformer related optimization, including BERT, GPT

A flexible and efficient deep neural network (DNN) compiler that generates high-performance executable from a DNN model description.

Library providing helpers for the Linux kernel io_uring support

A performant and modular runtime for TensorFlow

PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT

Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training

Simple Training and Deployment of Fast End-to-End Binary Networks

算法竞赛课件分享

Tensors and Dynamic neural networks in Python with strong GPU acceleration

The Tensor Algebra SuperOptimizer for Deep Learning

darktable is an open source photography workflow application and raw developer

Productive, portable, and performant GPU programming in Python.

HugeCTR is a high efficiency GPU framework designed for Click-Through-Rate (CTR) estimating training