AMD ROCm documentation

2026-04-08

3 min read time

Applies to Linux and Windows

ROCm is an open-source software platform optimized to extract HPC and AI workload performance from AMD Instinct GPUs and AMD Radeon GPUs while maintaining compatibility with industry software frameworks. For more information, see What is ROCm?

ROCm supports multiple programming languages and programming interfaces such as HIP, OpenCL, and OpenMP, as explained in the Programming guide.

If you’re using AMD Radeon GPUs or Ryzen APUs in a workstation setting with a display connected, review ROCm on Radeon and Ryzen documentation.

Note

The AMD ROCm Programming Guide presents key ROCm concepts in a structured, book-style format, a helpful starting point for those new to GPU programming.

ROCm documentation is organized into the following categories:

Install

• ROCm on Linux

• HIP SDK on Windows

• ROCm on Radeon and Ryzen

• Deep learning frameworks

• Build from source

How to

• Use ROCm for AI

• AI tutorials

• Use ROCm for HPC

• System optimization

• AMD Instinct MI300X performance validation and tuning

• System debugging

• Use advanced compiler features

• Set the number of CUs

• Troubleshoot BAR access limitation

• ROCm examples

Conceptual

• GPU architecture overview

• File structure (Linux FHS)

• GPU isolation techniques

• Using CMake

• Inception v3 with PyTorch

Reference

• ROCm libraries

• ROCm tools, compilers, and runtime API

• GPU hardware specifications

• Hardware atomics operation support

• Environment variables

• Data types and precision support

• Graph safe support

• ROCm glossary