This week, NVIDIA and its partners are excited to announce the availability of DGX Spark, the world’s smallest AI supercomputer, delivering NVIDIA’s AI stack in a compact desktop form factor.
Built on the NVIDIA Grace Blackwell architecture, DGX Spark integrates NVIDIA GPUs, CPUs, networking, CUDA libraries and NVIDIA AI software, accelerating agentic and physical AI development.
What’s new with DGX Spark:
-
NVIDIA Grace Blackwell architecture with integrated GPU and CPU
-
20-core Arm processor with high-performance cores
-
128 GB unified system memory
-
Compact desktop form factor (150mm x 150mm x 50.5mm)
-
Advanced connectivity including Wi-Fi 7, 10 GbE, and ConnectX-7
-
Support for AI models up to 200 billion parameters (or 405B for dual-Spark configuration)
-
NVIDIA Sync, a system tray utility that provides a simple way of accessing your DGX Spark system from another machine when it’s running as a headless appliance
-
DGX Dashboard, a built-in dashboard that provides an overview of the system’s current operational metrics, the ability to apply updates, change some system settings, and access local Jupyter Notebooks.
-
A wide range playbooks, practical guides, to help you get started
-
NVIDIA Container Toolkit ready to use out of the box for running GPU-accelerated containers
For full details, please refer to the DGX Spark User Guide.
To purchase a founders edition DGX Spark, head to the marketplace.
To get started with your DGX Spark, check out the Playbooks.