Getting Started Guide: Using AMD ROCm™ Software on Radeon™ GPUs
Overview
Developers, researchers, and software engineers working with advanced AI models can leverage select AMD Radeon™ GPUs to build a local, private, and cost-effective solution for AI development. AMD supports RDNA™ 3 architecture-based GPUs for desktop-based AI workflows using AMD ROCm™ software on Linux and WSL 2 (Windows® Subsystem for Linux) systems.
Accelerate Machine Learning Workflows on Your Desktop
- A local PC or workstation with one or multiple high-end AMD Radeon™ 7000 series GPUs presents a powerful, yet affordable solution to address the growing challenges in AI development thanks to very large GPU memory sizes of 24GB, 32GB and even 48GB.
- Radeon™ 7000 series GPUs feature more than 2x higher AI performance per Compute Unit (CU) compared to the previous generation1, and are equipped with up to 192 AI accelerators.
Open-source Community Support and Collaboration
- AMD ROCm software allows developers the freedom to customize and tailor their GPU software for their own needs encouraging community collaboration.
- Support for Hugging Face models and tools on Radeon GPUs using ROCm, allowing users to unlock the full potential of LLMs on their desktop systems.
Resources
ROCm Support
Visit the Documentation Portal to get started training ML models on your local desktop.
Installation Instructions
Run the following commands:
sudo apt update
wget https://repo.radeon.com/amdgpu-install/6.1.3/ubuntu/jammy/amdgpu-install_6.1.60103-1_all.deb
sudo apt install ./amdgpu-install_6.1.60103-1_all.deb
sudo amdgpu-install -y --usecase=graphics,rocm
sudo usermod -a -G render,video $LOGNAME
Please review the output carefully, then reboot for the changes to take effect.
What’s New
Community Feedback Tracker for ROCm on Radeon
To support the community, AMD has created a tracker to capture feedback and provide status updates. Our aim is to work more closely with the community, including providing timelines in the next couple of weeks for open sourcing additional portions of our software stack and more hardware documentation.
Sign Up for ROCm News
Sign up for the ROCm newsletter to keep up-to-date on the latest releases and tips.
- Based on AMD internal measurements, November 2022, comparing the Radeon RX 7900 XTX at 2.5GHz boost clock with 96 CUs issuing 2X the Bfloat16 math operations per clocks vs. the RX 6900 XT GPU at 2.25 GHz boost clock and 80 CUs issue 1X the Bfloat16 math operations per clock. RX-821.
- Supported GPUs include the latest high-end AMD Radeon™ 7000 series desktop GPUs, including the Radeon RX 7900 XT/XTX and Radeon RX 7900 GRE, as well as the Radeon PRO W7800, Radeon PRO W7900 and the new Radeon PRO W7900 Dual Slot.
© 2023 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, AMD ROCm, RDNA, Radeon, and combinations thereof are trademarks of Advanced Micro Devices, Inc. Linux® is the registered trademark of Linus Torvalds in the U.S. and other countries. Microsoft and Windows are registered trademarks of Microsoft Corporation in the US and/or other countries. PyTorch, the PyTorch logo and any related marks are trademarks of The Linux Foundation. Ubuntu and the Ubuntu logo are registered trademarks of Canonical Ltd. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.