Notepad++ official repository

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

Composable Kernel: Performance Portable Programming Model for Machine Learning Tensor Operators

A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")

A fast communication-overlapping library for tensor parallelism on GPUs.

chipStar is a tool for compiling and running HIP/CUDA on SPIR-V via OpenCL or Level Zero APIs.

Fast Matrix Multiplications for Lookup Table-Quantized LLMs

Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

Online compiler for HIP and NVIDIA® CUDA® code to WebGPU

ROCm BLAS marshalling library

AMD SMI

AMD’s C++ library for accelerating tensor primitives