IoT Worlds
Local AI Development
AI PCArtificial Intelligence

Revolutionizing Local AI Development with Windows Copilot Runtime and Copilot+ PCs

Microsoft has unveiled groundbreaking advancements in AI-powered computing (AI PC) at the Build developers conference enabling local AI development, introducing Windows Copilot Runtime and Copilot+ PCs. These revolutionary technologies aim to empower developers to create cutting-edge local AI experiences and transform the Windows ecosystem. With AI infused at every layer, Copilot+ PCs and Windows Copilot Runtime promise unparalleled performance, efficiency, and innovation.

Unleashing the Power of Copilot+ PCs for Local AI Development

Copilot+ PCs represent a quantum leap in AI computing, boasting the world’s most powerful PC Neural Processing Units (NPUs) capable of delivering over 40 TOPS of compute. These PCs offer up to 20 times more power and 100 times greater efficiency compared to traditional PCs when running AI workloads. The NPU, integrated into a new System on Chip (SoC), enables outstanding performance, all-day battery life, and exceptional app experiences. Qualcomm’s Snapdragon X Series processors will be the first to power Copilot+ PCs, with Intel and AMD devices joining later this year.

Developers can leverage the Snapdragon Dev Kit for Windows, featuring a special developer edition Snapdragon X Elite SoC, to maximize their development potential. This kit offers a stackable form factor, impressive specifications, and flexibility to meet developers’ needs. Pre-order is open, contact us to develop together the most innovative AI apps!

Enabling Breakthrough AI Experiences with Windows Copilot Runtime

Windows Copilot Runtime is an end-to-end AI platform that integrates AI into every layer of Windows, revolutionizing the OS itself. It offers developers a range of options, from high-level APIs to bringing their own machine learning models. The Windows Copilot Library, a set of APIs powered by 40+ on-device AI models, empowers developers to create innovative AI experiences. AI frameworks like DirectML, ONNX Runtime, PyTorch, and WebNN, along with toolchains, further facilitate AI development across the Windows hardware ecosystem.

Microsoft has built remarkable first-party experiences using the Windows Copilot Runtime, such as Recall for instant search, Cocreator for collaborative image generation, and more. Third-party developers like Davinci Resolve, WhatsApp, and Luminar Neo are also leveraging the NPU to deliver AI experiences with reduced latency, faster task completion, enhanced privacy, and lower cloud compute costs.

Harnessing the Potential of Windows Semantic Index and Vector Embeddings API

Windows Semantic Index, a new OS capability, redefines search on Windows and powers experiences like Recall. It organizes vast amounts of information from text to images and videos, transforming data into a vector store accessible through natural language search. The semantic index is stored entirely on the user’s local device, ensuring privacy.

The Vector Embeddings and RAG API, available in the Windows Copilot Library, enables developers to build their own semantic index store with app data and bring natural language search capabilities to their apps. This exemplifies how Microsoft is building new features using the models and APIs in Windows Copilot Runtime and offering the same capability to developers.

Embracing Open Standards and Collaboration

Windows supports a diverse silicon ecosystem, simplifying model optimization across hardware with DirectML. Native PyTorch support on Windows through DirectML allows thousands of Hugging Face models to work seamlessly, with NPU support coming soon. Microsoft is collaborating with Nvidia to scale these development workflows to over RTX AI GPUs. Interested to obtain great discounts on AI GPUs? Contact us today!

The WebNN Developer Preview brings local AI scale to web apps, enabling developers to leverage hardware for AI experiences at near-native performance. ONNX Runtime with DirectML optimizes performance for generative AI models, allowing developers to extend their Windows applications to other platforms.

Additional details from Pavan Davaluri (Microsoft Corporate Vice President Windows and Devices), here!

Related Articles

WP Radio
WP Radio