ROCm

AMD ROCm™ Software - GitHub Home

HIP: C++ Heterogeneous-Compute Interface for Portability

Run LLMs on AMD Ryzen™ AI NPUs in minutes; purpose-built and deeply optimized for the AMD NPUs.

The HIP Environment and ROCm Kit - A lightweight open source build system for HIP and ROCm

[DEPRECATED] Moved to ROCm/rocm-libraries repo

HIPIFY: Convert CUDA to Portable C++ Code

AI Tensor Engine for ROCm

super repo for rocm systems projects

[DEPRECATED] Moved to ROCm/rocm-libraries repo. NOTE: develop branch is maintained as a read-only mirror

Dockerfiles for the various software layers defined in the ROCm software platform

This repository is a fork of the upstream AMDGPU Driver, used by the ROCm project. To submit a patch, please follow the guidelines at...

super repo for rocm libraries

HCC is an Open Source, Optimizing C++ Compiler for Heterogeneous Compute currently for the ROCm GPU Computing Platform

[DEPRECATED] Moved to ROCm/rocm-systems repo

[DEPRECATED] Moved to ROCm/rocm-libraries repo

Kubernetes (k8s) device plugin to enable registration of AMD GPU to a container cluster

Omnitrace: Application Profiling, Tracing, and Analysis

AMD's graph optimization engine.

A collection of examples for the ROCm software stack

FlyDSL is the Python front‑end of the project: a Flexible Layout Python DSL for expressing tiling, partitioning, data movement, and kernel structure at a high level.

[DEPRECATED] Moved to ROCm/rocm-systems repo

[DEPRECATED] Moved to ROCm/rocm-libraries repo

AOMP is an open source Clang/LLVM based compiler with added support for the OpenMP® API on Radeon™ GPUs. Use this repository for releases, issues, documentation, packaging, and examples.

AMD MIVisionX is a computer vision toolkit built around a highly optimized, conformant open-source implementation of the Khronos OpenVX™ 1.3.2 specification. As of the 4.0.0 release, MIVisionX...

Examples for HIP

AMD RAD's multi-GPU Triton-based framework for seamless multi-GPU programming

AiTer Optimized Model

[DEPRECATED] Moved to ROCm/rocm-libraries repo

Modular RDMA Interface

ROC System Management Interface

[DEPRECATED] Moved to ROCm/rocm-libraries repo

ROCm OpenOpenCL Runtime

GPUFORT: S2S translation tool for CUDA Fortran and Fortran+X in the spirit of hipify

[DEPRECATED] Moved to ROCm/rocm-systems repo

[DEPRECATED] Moved to ROCm/rocm-systems repo

[DEPRECATED] Moved to ROCm/rocm-libraries repo

[DEPRECATED] Moved to ROCm/rocm-systems repo

[DEPRECATED] Moved to ROCm/rocm-systems repo

A tool for generating information about the matrix multiplication instructions in AMD Radeon™ and AMD Instinct™ accelerators

[DEPRECATED] Moved to ROCm/rocm-libraries repo

repo for AMD ROCDXG project

spur your compute

[DEPRECATED] Moved to ROCm/rocm-libraries repo

AMD SMI

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

[DEPRECATED] Moved to ROCm/rocm-libraries repo

(Deprecated) hipCaffe: the HIP port of Caffe

[DEPRECATED] Moved to ROCm/rocm-libraries repo

[DEPRECATED] Moved to ROCm/rocm-libraries repo

[DEPRECATED] Moved to ROCm/rocm-libraries repo

Repository to host ROCm Developer Hub Notebook Tutorials

LLVM AMDGPU Assembler Helper Tools

A system validation and diagnostics tool for monitoring, stress testing, detecting, and troubleshooting issues impacting AMD GPUs in high-performance computing environments

Next generation library for iterative sparse solvers for ROCm platform

Ahead of Time (AOT) Triton Math Library

ROCm Device Libraries

Fortran interfaces for ROCm libraries

[DEPRECATED] Moved to ROCm/rocm-systems repo

ROCm's Thunk Interface

Bandwidth test for ROCm

[DEPRECATED] Moved to ROCm/rocm-systems repo

[DEPRECATED] Moved to ROCm/rocm-libraries repo

TransferBench is a utility capable of benchmarking simultaneous copies between user-specified devices (CPUs/GPUs)

This is ROCgdb, the ROCm source-level debugger for Linux, based on GDB, the GNU source-level debugger.

ROCprof Compute Viewer (RCV) is a tool for visualizing and analyzing GPU thread trace data collected using rocprofv3

CMake modules used within the ROCm libraries

High Performance Linpack for Next-Generation AMD HPC Accelerators

AMD ROCm Performance Primitives (RPP) library is a comprehensive high-performance computer vision library for AMD processors with HIP/OpenCL/CPU back-ends.

HRX: Hip Runtime Extended

A ROCm library for GPU-Initiated IO. This provides support for initiating IO from a ROCm-capable GPU against a range of targets including RDMA NICs, NVMe SSDs and AMD GPU DMA engines (SDMAs).

FLM app mirror

Device Metrics Exporter exports metrics from AMD devices (GPUs) to collectors like Prometheus.

A lightweight triton-based General Matrix Multiplication (GEMM) library.

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...

[DEPRECATED] Moved to ROCm/rocm-libraries repo

[DEPRECATED] Moved to ROCm/rocm-libraries repo

MIOpenGEMM is now deprecated

A practical guide to high-performance gluon kernel development on AMD GFX9 GPUs.

[DEPRECATED] Moved to ROCm/rocm-libraries repo

ROCm CLI is a command-line tool for setting up and running local AI on AMD GPUs, with a full-screen TUI dashboard for GPU telemetry, model serving, and chat.

AMD’s C++ library for accelerating tensor primitives