Microsoft and Nvidia Expand Local AI Processing on Windows PCs with New Hardware and Model-Routing Capabilities

Microsoft and Nvidia are enhancing Windows PCs with expanded local AI processing capabilities, allowing developers and businesses to run larger AI models and autonomous agents directly on devices without full reliance on cloud services. This initiative, announced at a Windows and Surface event, introduces new AI hardware like Nvidia's RTX Spark platform, which combines a Blackwell RTX GPU with a Grace CPU, supporting up to 128GB of unified memory. Microsoft is also extending GitHub's HydraFusion technology to route tasks between cloud-based and local models, with experimental support for GitHub Copilot and Visual Studio Code planned for later in October. The approach, termed 'hybrid intelligence,' aims to optimize cloud computing budgets while maintaining access to advanced AI.