AMD Radeon AI

The Next Frontier of AI

With AMD Radeon™, users can harness the power of on-device AI processing to unlock new experiences and gain access to personalized and fast AI performance.

AI Accelerators

AMD RDNA™ 4 architecture delivers over 4X more AI compute than the previous generation AMD RDNA 3 architecture. Featuring 2 AI Accelerators per compute unit, support for new data types, and Sparsity – Radeon RX 9000 Series Graphics cards deliver up to 1557 TOPs of performance.

Large VRAM

Radeon RX 9000 Series GPUs feature up to 16GB of GDDR6 to run large AI models. The Radeon RX 9070 with 16GB of video memory offers a great entry-point to AI acceleration on AMD Radeon Graphics.

Wide Compatibility

Radeon RX Graphics cards are programmable to support the major ML frameworks including Microsoft DirectML, and select Radeon RX Graphics cards also support the AMD ROCm™* open software platform.

Amuse AI

Easy AI with Amuse and AMD Radeon™ RX Graphics

Imagine, create, and transform with ease. Explore the vast world of generative AI with Amuse 2.0, a simple and powerful application optimized for AMD Radeon™ RX graphics cards. 

Download Amuse AI 

Preserving old memories - AMD Radeon™ and AI

Relive your most cherished memories today with AMD Radeon™ and AI

Watch Video 

Perfecting special moments - AMD Radeon™ and AI

Today, remove unwanted objects and perfect your most special memories with the power of AMD Radeon™ and AI

Watch Video 

Bring old photos back to life - AMD Radeon™ and AI

Experience how, with the power of AMD Radeon™ and AI, you can bring old photos back to life today.

Watch Video 

AMD Ryzen Radeon laptop

How to run a LLM on your AMD Ryzen™ AI PC or Radeon Graphics Card

Did you know that you can run your very own instance of a GPT based LLM-powered AI chatbot on your Ryzen™ AI PC or Radeon™ 7000 series graphics card?

Read the Blog 

AMD Radeon 7000 Series

How to enable RAG on a AMD Ryzen™ AI PC or Radeon™ Graphics Card

Retrieval augmented generation (RAG) is a technique that allows you to enhance and provide context to your personal instance of a Large Language Model (LLM).

Read the Blog 

AMD ROCm

AMD Expands AI Offering for Machine Learning Development with AMD ROCm 6.0 and AMD Radeon™ GPUs

Building on our previously announced support of the AMD Radeon™ RX 7900 XT, XTX and Radeon PRO W7900 GPUs with AMD ROCm 5.7 and PyTorch, we are now expanding our client-based ML Development offering, both from the hardware and software side with AMD ROCm 6.0.

Read the Blog 

Applications

Today, we see applications emerging that use GPU hardware acceleration for AI workloads — including general AI compute, gaming/streaming, content creation and advanced machine learning model development. With the latest updates to the AMD RDNA 4 architecture, Radeon 9000 Series graphics cards are designed to accelerate AI in several use cases. 

Man playing video game on AMD computer

Gaming & Streaming

Elevate the streaming experience using AI-accelerated quality enhancements in the media engine and intelligent background noise removal through AMD Noise Suppression.

Learn About AMD Noise Suppression
Abstract red dots

AI and AMD Software

Experience all new AI-powered features such as AMD Chat, right within AMD Software.

Learn About AMD Software
Man working on an AMD laptop

Generative AI

Radeon™ graphics are ready to handle the most popular generative AI models.

Asian man using AMD laptop for content creation

Content Creation

Enhance content creation workflows and asset quality in popular apps such as Adobe® Photoshop® software , DaVinci Resolve and Blender®.

Learn About Radeon for Creators
Abstract 3d illustration - glowing blue and orange blocks of different sizes

ML Development

Develop advanced ML models by leveraging the dedicated AI hardware and memory subsystem on Radeon 9000 Series graphics cards for local, private, low-latency and low-cost ML training.

Learn About ML Development

Explore AI on Radeon Today

Explore how easy it is to run powerful offline AI workloads on the latest AMD Radeon RX 9000 Series graphics with the most popular applications and models.

Generative AI (Visual)

App Description
Amuse  Easy to install implementation for Stable Diffusion with an easy to use UI, and direct download of popular models from HuggingFace.com.
Automatic1111  Most popular PyTorch implementation of Stable Diffusion.
Stable Diffusion WebUI - DirectML  Optimized python version of Stable Diffusion for Radeon GPUs using DirectML.

Generative AI (LLM)

App Description
LM Studio  Easy install, fully accelerated offline application.
OLLAMA  Optimized for AMD Radeon to run Llama 2, Code Llama, and other models.

Creator

App Description
Adobe Creative Cloud   Adobe Premiere, Photoshop, Lightroom & After Effects from Adobe featuring online (Firefly) and offline (Neural Filters) AI features.
Black Magic Davinci Resolve Studio 18.6   Editing, color correction, visual effects, motion graphics and audio post production all in one software tool, with powerful Neural Engine acceleration via DirectML (DaVinci Resolve Studio only).
Topaz Labs   AI powered Photo and Video enhancement software for professionals and optimized for Radeon.
Black fade

Radeon Product Portfolio

Footnotes

Based on specifications of AMD RDNA 4 architecture compared to AMD RDNA 3 architecture as of December 2024. RX-1143

Radeon™ AI technology is compatible with all AMD Radeon™ 7000 Series graphics cards and newer. Please check with your system manufacturer for feature availability prior to purchase. GD-232.

PyTorch, the PyTorch logo and any related marks are trademarks of The Linux Foundation.

*Available on Radeon RX 7900 XTX and Radeon Pro W7900 in Linux