Nvidia just made the clearest case yet that the future of personal computing runs through AI hardware. At GTC Taipei during Computex, the company and Microsoft jointly revealed the RTX Spark, an Arm-based superchip designed to cram data-center-level AI performance into laptops thin enough to slide into a backpack.
The chip delivers up to 1 petaflop of FP4 AI performance.
What’s actually inside the RTX Spark
The RTX Spark pairs a Blackwell RTX GPU with up to 6,144 CUDA cores alongside a 20-core Arm-based Grace CPU. The chip includes 128 GB of unified memory, meaning the CPU and GPU share the same pool of RAM rather than shuttling data back and forth between separate banks.
Nvidia is emphasizing power efficiency as a core selling point. The design targets all-day battery life in thin-and-light form factors.
