Nvidia and Microsoft Unveil RTX Spark Superchip for AI-Native PCs
Nvidia and Microsoft have announced a new PC platform built around the RTX Spark Superchip — an Arm-based SoC combining GPU, CPU, and unified memory, much like Apple’s M-series chips — which they describe as a “new chapter” for personal computing. Laptops and desktops will ship with up to 128 GB of unified memory, enough to run large language models of up to 120 billion parameters locally in FP4 precision. The vision is that every PC gains a persistent local AI assistant capable of accessing Windows application windows, with deep OS-level security controls for data access. The platform aims to make on-device AI agents like OpenClaw or Hermes a standard part of the PC experience, though real-world adoption will depend on chip iterations, developer adaptation, and integration maturity over the coming years.