Urgent.News

What's breaking now, across thousands of outlets.

AI

NVIDIA DGX Spark Developer Kit Turns a Desk Into a Supercomputing Lab

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 […]

NVIDIA DGX Spark Developer Kit Turns a Desk Into a Supercomputing Lab

At the SC26 supercomputing conference in Chicago, NVIDIA unveiled the DGX Spark HPC Developer Kit designed for researchers, academics, and developers. This kit consists of a four-node cluster of NVIDIA DGX Spark systems, each directly connected to its neighboring nodes via NVIDIA ConnectX-7 networking. The kit provides researchers with a powerful workstation, enabling them to run multi-node scientific workloads, AI models, and agents without relying on shared GPU clusters.

The DGX Spark Developer Kit features the GB10 Grace Blackwell Superchip, delivering 512GB of combined memory and 20TB of NVMe storage. With NVFP4, researchers can deploy models with up to 700 billion parameters locally across the four DGX Spark clusters, such as DeepSeek V4.1. This allows teams to develop and debug their applications on their own, saving server hours for production runs and keeping sensitive datasets on-premises.

The NVIDIA HPC SDK playbook will soon be available, providing GPU-accelerated development tools such as compilers, libraries, and message passing interface tools. This enables researchers to build and test scientific applications across multiple nodes, optimize data distribution, and debug communication issues before scaling to larger systems.

The kit has already been validated by various scientific simulations and AI models, including materials simulation, molecular dynamics, weather modeling, and protein structure prediction.

The kit is available for purchase from top OEM partners and through the NVIDIA Marketplace. An NVIDIA playbook for building and running the NVIDIA HPC SDK will soon be released, allowing scientists and researchers to develop multi-node GPU applications and explore FP32 and mixed-precision algorithms before scaling to data center systems.

Written by urgent.news from NVIDIA Blog's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at blogs.nvidia.com →

More in AI

More from Monday 16 November →