The Future of AI on Windows 11 PCs: Supporting Windows ML to Scale AI Experiences

May 19, 2025

Join us at Microsoft Build

AI is set to revolutionize the way we work, entertain, and plan our daily lives on Windows 11 PCs. With the ability to handle complex AI workloads locally, Copilot+ PCs offer unprecedented performance, efficiency, and enhanced privacy, transforming various aspects of our digital experiences. AMD is at the forefront of the Copilot+ PC transformation, and with the Ryzen™ AI 3001 Series processor launched last year, we are now in our third generation of APUs with dedicated NPUs. Only AMD offers the broadest and most performant Copilot+ PC portfolio2, that ranges from ultrathin business systems to powerful gaming notebooks and even to everyday consumer notebooks.

Key features of the AMD Ryzen™ AI 300 Series processor include best-in-class multi-threaded CPU performance to get more work done in less time, a leadership NPU with 50+ TOPS for delivering enhanced AI experiences3, all-day battery life4 for on-the-go usage, a dedicated graphics engine for immersive gaming and content creation, and optimized, reliable drivers for seamless support of Windows 11.

With the introduction of the AMD Ryzen ™ AI Max+ processor, AMD continues to push the boundaries of PC hardware. AMD Ryzen ™ AI Max+ processors create a new class of PCs with up to 112GB of system memory dedicated for the GPU to execute AI tasks.  It can run massive LLM and image generation models locally, without the slowdown of page-to-RAM required for discrete graphics. LLMs like Llama 70B and Deepseek run faster and at lower power than the NVIDIA RTX 5080 discrete GPU. Cloud-quality models like Stable Diffusion 3.5 Large and Flux can be run completely locally and in FP16 for better accuracy than int8. This is unprecedented AI performance and efficiency delivered in a mobile form factor for AI enthusiasts, developers, and content creators.

Collaboration with PC Software Developer Ecosystem

As AMD continues to push the boundaries of Windows PC hardware with the Ryzen™ AI 300 Series and Ryzen™ AI Max+ processors, the collaboration with the PC software developer ecosystem is critical for a great user experience. AI developers often find themselves overwhelmed by the myriad of AI models, hardware accelerators, and optimization tools available. They need to constantly spend time learning different tools and optimizing for various accelerators. This complexity can detract from their primary focus, which is developing cutting-edge AI applications.  To address this AMD has partnered with Microsoft to revolutionize the way ISVs develop and deploy AI experiences across various IHV platforms with a new stack called Windows ML, powered by ONNX Runtime, for all Windows 11 PCs.

The Windows ML promise of "write once, run anywhere" simplifies the deployment process for app developers, allowing them to focus on delivering innovative user experiences instead of different IHV tools and optimizing individually for all the various hardware accelerators available. This approach resonates with the developer community, as it enables them to scale AI across CPUs, GPUs, and NPUs seamlessly including the latest AMD Ryzen™ AI 300 Series APU products, providing the best balance of performance and battery life. The adoption of the ONNX Runtime and the standard ONNX model format further lowers the bar for ISVs to onboard their workloads, making AI development on Windows more accessible and efficient while enhancing the overall user experience.

The automatic installation of Windows ML runtime and EPs through the Microsoft Store on Windows ensures that app developers and end users always have access to the latest, high-quality code components in the Windows ML stack, eliminating the need to include them directly in their apps. This standardized development platform reduces fragmentation in the Windows PC landscape, accelerating the adoption of AI and bringing new, engaging experiences to Windows PC clients. The deep partnership between AMD, Microsoft, and AI app developers, has been marked by a strong sense of co-engineering and mutual support, driving the pace of innovation and delivering great app experiences for the Windows ecosystem.

Support for Windows ML Stack

AMD is pleased to fully support the Windows ML stack for developer usage on our Ryzen™ AI 300 Series processors, including both Windows inbox models and Bring Your Own Models (BYOM). AMD unlocks the power of BYOM Windows ML development via the AMD Ryzen™ AI SDK v1.5 release available in late June. Find out more at AMD Ryzen™ AI Software.

End users should be on the lookout for exciting Windows applications utilizing Windows ML coming soon from our ISV partners such as BUFFERZONEPowderReincubateTopaz Labs, and more. If you're a developer, head over to Windows ML site to get started developing for Windows ML today!     

Disclaimers: 

  1. GD 220 AMD Ryzen™ AI is defined as the combination of a dedicated AI engine, AMD Radeon™ graphics engine, and Ryzen™ processor cores that enable AI capabilities. OEM and ISV enablement is required, and certain AI features may not yet be optimized for AMD  Ryzen AI processors. AMD Ryzen AI is compatible with: (a) Ryzen 7040 and 8040 Series processors and Ryzen™ PRO 7040/8040 Series processors except Ryzen 5 7540U, Ryzen 5 8540U, Ryzen 3 7440U, and Ryzen 3 8440U processors; (b) AMD Ryzen AI 300 Series processors and AMD Ryzen™ AI PRO 300 Series processors; (c) all AMD Ryzen 8000G Series desktop processors except the Ryzen 5 8500G/GE and Ryzen 3 8300G/GE; (d) AMD Ryzen 200 Series processors and Ryzen PRO 200 Series processors except Ryzen 5 220 and Ryzen 3 210; and (e) AMD Ryzen AI Max Series processors and Ryzen AI PRO Max Series processors. Please check with your system manufacturer for feature availability prior to purchase. GD-220e.

  2. KRKP 9: As of January 2025, AMD offers the most comprehensive lineup of next-generation mobile PC processors for Enterprise PCs supporting Copilot+ PCs among all PC processor manufacturers. 

  3. Testing as of Sept 2024 by AMD performance labs using the following benchmarks: Blender, Cinebench R24, Geekbench 6.3, and Passmark 11, systems: HP EliteBook X G1a with AMD Ryzen AI 9 HX PRO 375 processor @54W, Radeon 880M graphics, 32GB of RAM, 512GB SSD; Lenovo ThinkPad T14s Gen 6 with AMD Ryzen™ AI 7 PRO 360 processor @22W, Radeon™ 880M graphics, 32GB RAM, 1TB SSD; Dell Latitude 7450 with Intel Core Ultra 7 165U processor @15W (vPro enabled), Intel Iris Xe Graphics, 32GB RAM, 512GB NVMe SSD; Dell Latitude 7450 with Intel Core Ultra 7 165H processor @28W (vPro enabled), Intel Iris Xe Graphics, 16GB RAM, 512GB NVMe SSD,. All systems Windows 11 Pro, VBS=ON, and tested in Best Performance Mode. PassMark is a registered trademark of PassMark Software Pty Ltd. AI PC is defined as a laptop PC with a processor that includes a neural processing unit (NPU). STXP-07.

  4. GD 173a : AMD defines “All Day Battery Life” as at least 8 hours of continuous battery life and "Multi-Day battery Life" as continuous runtime above 8 hours. All battery life scores are approximate. Actual battery life will vary based on several factors, including, but not limited to: system configuration and software, settings, product use and age, and operating conditions. GD-173a.  

Share:

Article By


Corp. VP, Product Management

Related Blogs