AI hardware and software developer NVIDIA announced all-new personal AI supercomputers. Powered by the NVIDIA Grace Blackwell platform, the DGX supercomputers enable users to create, refine, and run large AI models on their personal computers.
AI Supercomputer
DGX Spark—formerly Project DIGITS—and DGX Station are designed for AI-native developers to run AI-native applications, which are AI integrated into the core design rather than an add-on.
“AI has transformed every layer of the computing stack. It stands to reason a new class of computers would emerge — designed for AI-native developers and to run AI-native applications,” said Jensen Huang, founder and CEO of NVIDIA. “With these new DGX personal AI computers, AI can span from cloud services to desktop and edge applications.”
DGX Spark
NVIDIA claims that the DGX Spark is the “world’s smallest AI supercomputer,” with “massive” generative and physical AI performance and capabilities. At the heart of the AI supercomputer is the NVIDIA GB10 Grace Blackwell Superchip, developed specifically for desktop form factors, which are the size, shape, and layout of computer components such as the motherboard.
The GB10 superchip features a powerful GPU, or graphics processing unit, that delivers “up to 1,000 trillion operations per second of AI compute for fine-tuning and inference with the latest AI reasoning models,” according to an NVIDIA press release.
DGX Station
According to NVIDIA, the DGX Station is the only desktop system with the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip, featuring 784GB of “coherent” memory space. With a large amount of memory space, it’s capable of accelerating large-scale training and inferencing workloads.
Reservations are open for the DGX Spark. NVIDIA expects the DGX Station to be available later this year from manufacturer partners such as ASUS, BOXX, Dell, HP, Lambda, and Supermicro.
