AMD ROCm™ Software

rocHPL

High Performance Linpack for Next-Generation AMD HPC Accelerators

rocBLAS-Examples

Examples illustrating usage of the rocBLAS library

rocm-spack

A flexible package manager that supports multiple versions, configurations, platforms, and compilers.

distributed

A distributed task scheduler for Dask

OpenLLM

Operating LLMs in production

nccl-rccl-parser

Tool to run rccl-tests/nccl-tests based on from an application

pytorch-lightning

The lightweight PyTorch wrapper for high-performance AI research. Scale your models, not the boilerplate.

MIFin

Tuna centric MIOpen client

miopen_cudnn_ops

pyrsmi

python package of rocm-smi-lib

pytorch-micro-benchmarking

shap

A game theoretic approach to explain the output of any machine learning model.

rocgputreeshap

ROCm support for GPUTreeShap

HIP-CPU

An implementation of HIP that works on CPUs, across OSes.

gloo

Collective communications library with various primitives for multi-machine training.

tensorpipe

A tensor-aware point-to-point communication primitive for machine learning

aws-ofi-rccl

rocGemmDriver

rocGemmDriver

atmi

Asynchronous Task and Memory Interface, or ATMI, is a runtime framework and programming model for heterogeneous CPU-GPU systems. It provides a consistent, declarative API to create task graphs on CPUs and GPUs (integrated and discrete).

FAMBench

Benchmarks to capture important workloads.

AITemplate

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.

og

OpenMP GCC support

ROCm-OpenCL-Runtime

ROCm OpenOpenCL Runtime

ROCclr

hipamd

rdma-perftest

rocm-recipes

Recipes for rocm

frugally-deep

Header-only library for using Keras (TensorFlow) models in C++.

roc-stdpar

ClassyVision

An end-to-end PyTorch framework for image and video classification