Windows ML + AMD: Unlocking AI for Every Windows Developer

As AI PCs reshape the future of computing, AMD delivers the hardware muscle while Windows ML – now in general availability - weaves a complex hardware accelerator ecosystem of CPUs, GPUs, and NPUs into a cohesive, abstracted whole. Developers can leverage AMD and Windows ML runtime with backend optimizations built in, allowing them to focus on building better local AI solutions.

Together, AMD and Microsoft empower a new era of intelligent applications, frictionless local inference, and easily deployable AI experiences across the Windows ecosystem as Windows ML enters general availability for local AI development.

What Is Windows ML and Why Developers Should Care

Windows ML is Microsoft’s machine learning API for Windows, built on the ONNX Runtime. It allows developers to run AI models locally on devices, automatically selecting the best execution provider—CPU, GPU, or NPU—based on available hardware capabilities.

For developers, this means:

• Faster time to market: Windows ML abstracts hardware complexity, freeing developers to focus on feature implementation and rapid iteration.
• Simplified Development: Integrate trained models directly in Windows apps, handling dependency management and hardware optimization behind the scenes.
• Privacy and security: AI inference runs safely and efficiently on the device itself.

Whether you’re building productivity tools, content creation software, or enterprise apps, Windows ML makes it easy to deploy and scale solutions.

How AMD Hardware Supercharges Windows ML

The latest AMD Ryzen™ AI, Ryzen™ AI Max, Ryzen™ AI PRO and Ryzen™ AI Max PRO Series processors deliver up to 55+ TOPS of dedicated AI compute via the integrated XDNA™ NPU.  The Ryzen™ AI Max and Ryzen™ AI Max PRO processors deliver higher performance and support up to 128GB of unified DRAM with double the memory bandwidth, capable of running models with over 100 billion parameters. A cornerstone of the approach from AMD is balance – all AMD Ryzen AI processors provide CPU, GPU, and NPU ML acceleration offering developers an efficient, flexible architecture that delivers great user experiences across a wide spectrum of real-world AI workloads.

AMD Ryzen AI combined with Windows ML provides:

• Scalable acceleration: Select the optimal compute—CPU, NPU, or GPU—to meet performance needs from lightweight inference to high-throughput workloads.
• Power efficiency: Choosing the right processing block for a task extends battery life in AI PCs without compromising performance. Windows ML supports auto-EP selection based on developer hints.
• Developer-first flexibility: Windows ML automatically routes work to the right AMD compute engine, with no code rewrites required.

At AMD, we are deeply committed to empowering developers and customers with the tools they need to unlock the full potential of AI. Our balanced approach spanning CPU, GPU, and NPUs delivers high performance when it matters most for applications and also energy efficient compute to power always-on features with minimal impact to battery life. Combined with support for Windows ML, this makes AI more accessible than ever before, enabling innovation across a wider range of people, devices, and use cases.

Strategic Impact for OEMs, ISVs, and Enterprises

Windows ML is not just a developer convenience, it is a strategic enabler for the entire Windows ecosystem.

• OEMs: Windows ML’ s runtime flexibility makes it easier to deliver Copilot+ PC experiences advanced by AMD Ryzen AI, Ryzen AI PRO, Ryzen AI Max and Ryzen AI Max PRO Series processors.
• ISVs: Build cross-platform AI apps that scale easily across systems, from mainstream laptops to desktop-equivalent mobile workstations.
• Enterprises: Deploy AI workflows with stronger privacy, low latency, and reduced cloud costs.

“Windows ML runs seamlessly across AMD CPUs, GPUs, and NPUs—including Ryzen AI 300 Series—enabling powerful AI experiences. Our deep partnership with AMD is driving the future of AI on Windows 11 and we’re excited it’s generally available for developers today.” — Logan Iyer, VP, Distinguished Engineer, Windows Platform & Developer

By aligning Windows ML with the AMD hardware roadmap, we help ensure that every AI-enabled Windows device can deliver consistent, excellent performance, both today and in the future.

Building the Future Together

At AMD, we believe in open standards, developer-first design, and scalable innovation. Our partnership with Microsoft on Windows ML reflects that commitment, enabling developers to build smarter, fast, more protected AI applications before deploying them across the Windows ecosystem.

Whether you’re optimizing existing projects, exploring new creative ideas, or enabling enterprise AI, AMD and Microsoft are here to help you build what’s next.

Ready to Get Started?

Explore these resources to begin building with AMD + Windows ML:

• AMD Developer Central
• ONNX Runtime
• For detailed tutorials, API reference, and sample code, visit the Microsoft Learn page for Windows ML

Let’s build the future together.