NVIDIA and Microsoft Bring AI Agents to Windows PCs with RTX Spark

Windows Becomes a Platform for AI Agents

At the Windows AI and Surface event in San Francisco on Wednesday, NVIDIA founder and CEO Jensen Huang and Microsoft CEO Satya Nadella outlined how the two companies are co-engineering hardware and software to run AI agents on Windows PCs. Speaking during a fireside chat hosted by Sriram Krishnan, a former senior White House policy advisor on AI, Huang traced NVIDIA's history back to Windows, saying "If not for Windows there would be no GeForce." Nadella credited Huang's long-term vision and described the personal computer as the ultimate tool, now becoming a personal assistant in the era of agents.

Microsoft announced general availability of Microsoft Execution Containers (MXC), OS-level infrastructure that lets agents run safely and persistently in the background under operating system control. Pavan Davuluri, Microsoft's EVP of Windows and Devices, said the primitives are built directly into Windows so agents can be secured, observed and governed, together with Microsoft Security and Agent 365. Huang compared the shift to earlier platform changes, saying MXC will revolutionize how agents are built and deployed, just as Windows and DirectX revolutionized applications.

RTX Spark Brings Local AI to Laptops and Compact Desktops

RTX Spark puts the full NVIDIA AI stack into Windows laptops and compact desktops, with laptop preorders opening at the event and availability on October 16, followed by compact desktops in November. The platform combines an NVIDIA Blackwell RTX GPU with up to 6,144 cores and an up to 20-core NVIDIA Grace CPU connected at 600 GB/s, delivering one petaflop of FP4 AI performance and up to 128GB of unified memory. According to NVIDIA, it can run models such as Qwen 3.8 Flash Next, a 125B model with 51B n-gram, locally and unmetered without sending data to the cloud.

Microsoft introduced the Surface Laptop Ultra, built around RTX Spark with up to 128GB of unified memory and up to a petaflop of AI compute. Systems are also coming from Acer, ASUS, Dell, HP, Lenovo, Microsoft, MSI and Gigabyte, ranging from slim laptops to compact desktops designed for always-on agents. RTX Spark runs the full NVIDIA CUDA platform, so developers can move models, tools and workflows from RTX Spark to DGX Station without rewriting. Creators get 5th-generation Tensor Cores with NVFP4 support, hardware-accelerated AV1 and 4:2:2 video encode and decode, DLSS and RTX ray tracing, while gamers can run AAA games at 1440p over 100 frames per second with DLSS 5, Reflex and G-SYNC.

DGX Station for Windows Targets Enterprise AI

The event also previewed NVIDIA DGX Station for Windows, described as the first deskside AI supercomputer to bring GB300 Grace Blackwell-class infrastructure into the Windows ecosystem. Until now, DGX Station ran on Linux, forcing enterprise developers to maintain separate Linux and Windows environments. The new system runs on the GB300 Grace Blackwell Ultra Desktop Superchip with 748GB of coherent memory and up to 20 petaFLOPS of FP4 AI compute, enough to run models up to a trillion-parameter scale locally. Linux AI toolchains remain accessible through WSL when needed.

Sources: Nvidia