NVIDIA Launches Four-Node DGX Spark HPC Developer Kit
NVIDIA announced the DGX Spark HPC Developer Kit at SC26, offering researchers and developers a four-node DGX Spark cluster connected in a ring with ConnectX-7 networking and no separate switch. NVIDIA says the system provides 512GB of combined memory and can deploy models up to 700B parameters locally with NVFP4, including DeepSeek V4.1. The kit is available for purchase through OEM partners and NVIDIA Marketplace, while an installation playbook for NVIDIA HPC SDK is still forthcoming.
The purchasable kit gives research teams a local platform for developing and debugging multi-node scientific applications, AI models, and agents before using larger shared systems.
At SC26, an annual supercomputing conference taking place this week in Chicago, NVIDIA is introducing the DGX Spark HPC Developer Kit for researchers, academics and developers.
The kit is a four-node cluster of NVIDIA DGX Spark systems, each connected directly to its two neighbors in a ring, with an NVIDIA HPC SDK installation playbook coming soon. Researchers get a place at their desk to run multi-node scientific workloads, AI models and agents, instead of waiting for time on a shared GPU cluster. The kit puts the GB10 Grace Blackwell Superchip on researchers’ desks.
A Supercomputing Lab on Any Desk
The DGX Spark Developer Kit changes how researchers work.
Read All NVIDIA Announcements From SC26
It scales from one system, or node, to four — linked directly to one another over NVIDIA ConnectX-7 networking, with no network switch to buy or set up.
#NVIDIA#DGX Spark HPC Developer Kit#DGX Spark#GB10 Grace Blackwell Superchip#ConnectX-7#NVIDIA HPC SDK#DeepSeek V4.1