NVIDIA and Microsoft Launch RTX Spark and AI Agents for PCs
At a recent Microsoft event in San Francisco, NVIDIA and Microsoft executives outlined a new era of co-engineered hardware and software designed to bring persistent AI agents to Windows PCs.
Radwan AbdulhadiEditor and technology writer
A New Era for Windows and AI Agents
At a Microsoft event held at Dogpatch Studios in San Francisco on Wednesday, NVIDIA founder and CEO Jensen Huang and Microsoft CEO Satya Nadella outlined a major collaborative push to bring AI agents directly to Windows PCs. During a fireside chat hosted by Sriram Krishnan, the executives discussed how co-engineering hardware and software will shape the future of personal computing.
Tracing back NVIDIA's multi-decade history with Microsoft, Huang noted that Windows has always been at the core of the company's journey, stating that without Windows there would be no GeForce. Nadella praised Huang's long-term vision regarding the secular trend of agentic AI, emphasizing the goal of making the desktop a secure execution environment for personal assistants.
OS-Level Security with Microsoft Execution Containers
To support always-on agents safely, Microsoft announced the general availability of Microsoft Execution Containers (MXC). Pavan Davuluri, Microsoft’s EVP of Windows and Devices, explained that MXC serves as the OS-level infrastructure enabling agents to run persistently and securely in the background under strict operating system governance.
According to Microsoft, primitives built directly into Windows ensure that agents can be observed, secured, and governed effectively. Jensen Huang compared this development to historical milestones like DirectX, predicting that MXC will similarly revolutionize how applications and agents are built and deployed.
Introducing NVIDIA RTX Spark for Laptops and Desktops
Among the hardware announcements, the new NVIDIA RTX Spark platform integrates the full NVIDIA AI stack into Windows laptops and compact desktops. Designed to run advanced local models without relying on the cloud, laptop preorders open today and units will be available starting October 16, while compact desktops are scheduled for release in November. Users can review purchasing options by visiting the NVIDIA Blog for official updates.
Pavan Davuluri highlighted the hardware integration with devices like the new Surface Laptop Ultra, which is built around NVIDIA RTX Spark. Offering up to 128GB of unified memory and up to a petaflop of AI compute, the system allows users to run complex models locally. Devices spanning various form factors are slated from major manufacturers including Acer, ASUS, Dell, HP, Lenovo, Microsoft, MSI, and Gigabyte.
Architecture and Capabilities of RTX Spark
The RTX Spark superchip pairs an NVIDIA Blackwell RTX GPU featuring up to 6,144 cores with an up to 20-core NVIDIA Grace CPU, connected via a 600 GB/s link. Delivering one petaflop of FP4 AI performance and up to 128GB of unified memory, the hardware efficiently handles local models such as Qwen 3.8 Flash Next, a 125B model with 51B n-gram that matches cloud-tier intelligence securely and unmetered.
Running the full NVIDIA CUDA platform, RTX Spark caters to multiple use cases across development, content creation, and gaming. Developers can seamlessly shift workflows to DGX Station, creators benefit from 5th-generation Tensor Cores with NVFP4 support alongside hardware-accelerated AV1 and ray tracing, and gamers can enjoy high frame rates at 1440p using DLSS 5, Reflex, and G-SYNC.
Enterprise AI Compute with NVIDIA DGX Station for Windows
Expanding local AI beyond individual consumer PCs, the event also previewed the NVIDIA DGX Station for Windows. This marks the first deskside AI supercomputer to bring GB300 Grace Blackwell-class infrastructure directly into the Windows ecosystem, effectively bridging the traditional gap between enterprise Windows productivity tools and heavy Linux-based AI workflows.
Driven by the GB300 Grace Blackwell Ultra Desktop Superchip, the DGX Station for Windows delivers 748GB of coherent memory and up to 20 petaFLOPS of FP4 AI compute. This performance allows development and research teams at Fortune 500 companies and labs to fine-tune and run multi-agent workloads and trillion-parameter models locally while remaining integrated with their existing enterprise applications.
Sources
- NVIDIA BlogNVIDIA, Microsoft Kick Off a New Beginning for Windows PCs With RTX Spark and AI Agents
Continue chronologically
Related entity coverage
Related stories
Microsoft Hosts Windows and Surface Event in San Francisco
Microsoft holds a major Windows and Surface event in San Francisco featuring leaders from Microsoft and NVIDIA to discuss local AI and new hardware.
Microsoft Prepares for Windows, Surface, and AI Event
Microsoft is holding a major event in San Francisco to discuss local AI, the Surface Laptop Ultra, and upcoming agentic AI software integrations.
Microsoft and NVIDIA Showcase Local AI and New Windows Features
Microsoft hosts an event in San Francisco featuring CEO Satya Nadella and NVIDIA CEO Jensen Huang to highlight local AI and new Windows capabilities.
Inside Microsoft's Lab to Build the Surface Laptop Ultra
Discover how Microsoft engineered its MacBook Pro competitor, featuring Nvidia's RTX Spark chip and rigorous robotics testing in Redmond.
