In our mission to bring AI everywhere, Intel is committed to investing in the AI ecosystem to ensure its platforms are ready for the industry’s newest AI models and can provide a frictionless experience for developers. Today, Microsoft announced the public preview for Windows* ML, enhanced by the built-in performance advantage of the OpenVINO™ toolkit on Intel hardware. This collaboration allows developers to seamlessly leverage AI inferencing capabilities in Windows* applications, marking a significant milestone to scale AI on Windows 11*.
Empowering Developers with OpenVINO™ Execution Provider for Windows* ML
In 2024, Intel and Microsoft joined forces and introduced Copilot+ PCs, with exclusive AI experiences, powered by Intel® Core™ Ultra 200V Series processors with integrated NPUs up to 48 TOPS. The Windows Inbox models powering these features can be accessed by developers through the Windows* App SDK, enabling them to incorporate these advanced capabilities into their own applications.
Intel and Microsoft are now extending this partnership to empower software developers through the OpenVINO™ Execution Provider for Windows* ML. Integrated with the Windows* ML API and distributed via the Microsoft Store on Windows*, this solution delivers continuous AI innovation and seamless hardware acceleration across Intel CPUs, GPUs and NPUs. Developers can focus on building and deploying AI applications faster, leveraging native Intel optimizations while reducing integration and deployment overhead.
Key Benefits:
- Performance at Scale: Harness the power of the OpenVINO™ toolkit with the AI Toolkit for Visual Studio Code (VS Code)* to produce optimal models with enhanced accuracy and performance.
- Hardware Flexibility: Utilize device selection capabilities for flexible HW choice between Intel CPU, GPU or NPU accelerators.
- Cost Efficiency: Reduce development overhead with pre-validated and optimized models accessible through streamlined Windows* APIs.
- Simplified Development: Leverage Windows* APIs for AI app development where Windows* ML takes care of all package dependency management.
- Enhanced Serviceability: Ensure seamless distribution of performance enhancements and new capabilities across Intel platforms and devices, eliminating the need for app updates.
For additional fine-grained control on Intel hardware at the edge or in the cloud, the native OpenVINO™ toolkit is available today.
Partnerships
“Microsoft has partnered with Intel to streamline the deployment process on Windows* 11 + PCs powered by Intel processors. Combining the ease of development and deployment of Windows* ML with the performance and efficiency of OpenVINO™, we will enable developers to rapidly create, deploy and maintain AI applications across the large Windows* and Intel install base.”
– Logan Iyer, Distinguished Engineer, Windows* Platform + Developer, Microsoft
Intel has a long history of deep app developers (ISVs) enabling engagements focused on developing optimized software to take advantage of the full capabilities, features and performance of Intel hardware.
Recently, Intel has been working together with Powder*, a developer using cutting-edge, local AI to transform gameplay into captivating stories, to bring the best out of Intel powered AI PCs and Copilot+ PCs by leveraging NPU during gaming. Powder* has integrated the Windows* ML API, unlocking the AI acceleration powered by OpenVINO™ toolkit and ensuring optimal and continually updated performance and functionality across platforms.
“OpenVINO™ and Windows* ML let us seamlessly unlock NPU acceleration in Powder's* real-time AI engine. That means creators get an instant, hardware-optimized experience - recording, analyzing and editing in seconds, directly on Intel AI PCs and Copilot+ PCs.”
– Barthélémy Kiss, CEO and Co-founder, Powder*
Intel has likewise teamed up with BUFFERZONE* Safe Workspace*, an endpoint security solution that prevents external attacks through Application Isolation. It employs NoCloud* AI technology to address threats beyond the container and is currently available in the market. BUFFERZONE* partnered with Intel to enable NoCloud* AI technology inference across Intel’s CPU, GPU, and NPU, leveraging ONNX* and OpenVINO™ GenAI, and in their early development efforts, saw very positive results using Windows* ML.
"At BUFFERZONE*, we see AI-powered PCs as the future of endpoints. Windows* ML, coming soon with BUFFERZONE* Safe Workspace*, simplifies integration for developers and eases AI adoption, reducing time to market for new AI usages. This will provide customers with more value by leveraging Intel® Core™ Ultra processors on their Copilot+ PCs and AI PCs."
-Dr. Ran Dubin, CTO, BufferZone*
Experience Windows* ML on Intel at Microsoft Build
At the Microsoft Build conference, Intel is dedicated to empowering developers by showcasing the latest advancements in AI technology through a diverse array of sessions and workshops:
- Breakout Session: Accelerate AI from PC to Cloud with Intel, Copilot+ PCs and Azure. Explore effortless AI deployment on Intel PCs with Windows AI Foundry.
- Hands-On AI PC Workshops: Join us for three interactive workshops to learn and experience Windows* ML tech on Intel AI PCs and Copilot+ PCs (Tues AM, Tues PM, Wed).
- Booth Demos: Meet AI Director, which improves AI/ML performance by maximizing the usage of all processing units (CPU, GPU, NPU). Learn about Powder*, a Windows* ML & OpenVINO™ enabled application that utilizes the NPU for AI gaming training.
- AI Developer Workflows with Windows* ML: Explore streamlined AI workflows with Windows* ML in an on-demand session.
- Explore Intel Partner Page
We look forward to meeting you at Microsoft Build and getting your feedback on Windows* ML + OpenVINO™!