Use ROCm on Radeon and Ryzen

Use ROCm on Radeon and Ryzen#

Unlock Local AI Development on Your AMD Hardware

Transform your AMD-powered system into a powerful and private machine learning workstation. With the latest ROCm™ software stack, you can now harness the full potential of high-end AMD Radeon™ GPUs and Ryzen™ APUs for your AI workflows on both Linux® and Windows®.

ROCm™ 6.4.4 expands support to the latest Radeon™ 9000 Series (RDNA™ 4) and select 7000 Series (RDNA™ 3) GPUs for Windows, and introduces initial support for Ryzen™ APUs, enabling cost-effective, local development and inference for researchers and engineers using Pytorch.

Expanded Platform Support#

Quickly see what’s supported on your system. ROCm™ 6.4.4 focuses on bringing PyTorch support to new platforms while maintaining robust support on our established Linux platform for Radeon GPUs.

Hardware

Operating System

Supported Frameworks

Status

Radeon™ GPUs (9000 & select 7000 Series)

Linux®

PyTorch, TensorFlow, JAX, ONNX

Fully Supported

Radeon™ GPUs (9000 & select 7000 Series)

Windows®

PyTorch

New: Preview Support

Ryzen™ APUs (AI Max 300 Series & select AI 300 Series)

Linux®, Windows®

PyTorch

New: Preview Support

Your Local, Private AI Powerhouse#

Modern AI models demand significant computational power and memory. A local workstation equipped with a Radeon™ GPU, featuring up to 48GB of VRAM, offers a secure and economical alternative to relying solely on cloud-based solutions. Furthermore, expanded support to Ryzen™ APUs, with up to 128GB of shared memory, allow even laptop users to develop ML workflows securely and efficiently.

Leadership in Open-Source GPU Programming#

ROCm is the leading open-source software foundation for GPU programming. The same ROCm™ stack that powers your desktop development on RDNA™ architecture GPUs also supports AMD Instinct™ accelerators on CDNA™ architecture in the datacenter. This unified platform creates a seamless migration path, allowing you to develop applications locally and deploy them at scale with confidence. This maximizes GPU hardware investments, facilitating the development, testing, and deployment of AI workloads, GPU-accelerated HPC, scientific computing, CAD, and other applications.

As a primarily open-source ecosystem, ROCm™ gives you the freedom to inspect, customize, and tailor the software stack to your specific needs, backed by a collaborative community of developers.

ROCm™ Key Capabilities#

Frequent ROCm™ releases introduce major platform expansions and key software improvements.

New Platform Updates:

  • Windows Preview: For the first time, PyTorch is officially supported on Radeon™ GPUs and Ryzen™ APUs (Preview).

  • APU Linux Support: Initial PyTorch support and increased ROCm functionality for Ryzen™ APUs (Preview).

Established Features & Framework Support (Radeon on Linux):

  • Pytorch and Tensorflow: Full and established support for both training and inference.

  • vLLM: Full support.

  • JAX: Support for JAX (inference only).

  • Llama.cpp: Supported for efficient inference.

  • FlashAttention-2: Backward pass enabled for more efficient training.

  • ONNX Runtime: Expanded support for INT8 and INT4 inference with MIGraphX.

  • Operating Systems: Continuing support for Red Hat Enterprise Linux (RHEL) 9.6 as well as Ubuntu.

  • See the full support matrices here

Get Started#

Ready to build? Find all the resources you need at the links below.

Contents