AMD Enables DeepSeek R1 Reasoning Models on Ryzen AI and Radeon GPUs

AMD Brings Advanced AI to Consumer Hardware

AMD has taken a significant step in democratizing AI technology by providing instructions for running DeepSeek's R1 distilled "reasoning" models on AMD Ryzen AI processors and Radeon graphics cards. This move allows users to harness the power of sophisticated AI models on their personal computers, marking a notable advancement in consumer AI capabilities 1

1

2

Understanding DeepSeek R1 Reasoning Models

DeepSeek R1 is a frontier "reasoning" model, recently distilled into smaller, highly capable versions. These models are designed to tackle complex tasks using chain-of-thought (CoT) reasoning. Unlike conventional large language models (LLMs) that provide immediate responses, CoT LLMs perform extensive reasoning before answering, improving their analytical capabilities at the cost of longer response times 1

1

2

Deployment on AMD Hardware

The deployment of DeepSeek R1 distilled models on AMD hardware is remarkably straightforward, thanks to LM Studio's one-click installer tailored for Ryzen AI. AMD has provided a comprehensive guide detailing the process, including how to tune the application for optimal performance on their hardware 2

2

3

Hardware Compatibility and Model Support

AMD's latest consumer hardware, including RX 7000 series desktop GPUs and select Ryzen CPUs with XDNA NPUs, are compatible with several LLMs using R1. The maximum supported LLM parameters vary based on memory capacity:

• RX 7600 XT to 7900 XT: Up to "DeepSeek-R1-Distill-Qwen-14B"
• RX 7900 XTX: Up to "DeepSeek-R1-Distill-Qwen-32B"
• RX 7600: Up to "DeepSeek-R1-Distill-Llama-8B"
• Ryzen 8040 and 7040 series mobile APUs: Up to "DeepSeek-R1-Distill-Llama-14B"
• Ryzen AI Max+ 395: Up to "DeepSeek-R1-Distill-Llama-70B" (with sufficient RAM) 3

  3

Related Stories

NVIDIA Accelerates DeepSeek-R1 Reasoning Models with GeForce RTX 50 Series and NIM Microservice

Microsoft Embraces DeepSeek R1: A New Chapter in AI Accessibility and Competition

AMD Unveils Stable Diffusion 3.0 Medium AI Model for Local Image Generation on Ryzen AI Laptops

DeepSeek R1: Open-Source AI Model Rivals Proprietary Giants in Reasoning and Cost-Efficiency

DeepSeek R1's Impact on the AI Landscape

DeepSeek R1 has garnered significant attention in the AI community due to its impressive efficiency. The model boasts 11 times lower computing costs compared to leading-edge models, achieved through extreme optimization levels and the use of Nvidia's Parallel Thread Execution (PTX) programming 3

3

Broader Implications

The availability of DeepSeek R1 on consumer hardware represents a significant democratization of AI technology. It allows individuals to run sophisticated AI models locally, potentially opening up new avenues for personal and professional applications of AI. This development also highlights the ongoing competition in the AI hardware space, with AMD positioning itself as a strong contender in providing AI capabilities to end-users 1

1

2

3