"Nvidia's tiniest Grace-Blackwell workstation is finally making its way to store shelves this week, the better part of a year after the GPU giant first teased the AI mini PC, then called Project Digits, at CES. Since rebranded as the DGX Spark, the roughly NUC-sized system pairs a Blackwell GPU capable of delivering up to a petaFLOP of sparse FP4 performance with 128 GB of unified system memory and 200 Gbps of high-speed networking."
"But with a price starting around $3,000, small doesn't mean cheap. Then again, it's not exactly aimed at mainstream PC buyers. The systems, which will also be available under various brand names from OEM partners, won't even come with Windows. A Copilot+ PC this is not. Instead, it ships with a custom spin of Ubuntu Linux. Spark is actually intended for AI and robotics developers, data scientists, and machine learning researchers."
"Powering the DGX Spark is the GB10 system-on-a-chip, which is essentially a miniaturized version of the Grace-Blackwell Superchips that power its flagship NVL72 rack systems. As we explored back at Hot Chips, the GB10 is composed of two compute dies connected at 600 GB/s via Nvidia's proprietary NVLink chip-to-chip interconnect tech. And, in case you're wondering, this same technology will eventually be used to mesh Nvidia's GPUs to Intel's future client CPUs as part o"
Nvidia has launched the DGX Spark, a NUC-sized Grace-Blackwell workstation delivering up to a petaFLOP of sparse FP4 compute with 128 GB of unified system memory and 200 Gbps networking. The system starts around $3,000 and targets AI and robotics developers, data scientists, and machine learning researchers rather than mainstream PC buyers. Devices ship with a custom Ubuntu Linux spin rather than Windows. The Spark can run models up to 200 billion parameters and serves as a lower-cost, high-memory workstation alternative to expensive high-end GPU cards. The GB10 SoC uses two compute dies linked at 600 GB/s via NVLink.
Read at Theregister
Unable to calculate read time
Collection
[
|
...
]