Reported by 2 outlets
The story so far
AI summary of the reporting below · Updated · How we work
Nvidia has expanded the clustering capabilities of its DGX Spark platform, enabling users to link up to eight systems together. The expanded configuration doubles the previous four-way limit, allowing hardware setups to reach 512 GB of unified memory using 64 GB models or 1 TB using 128 GB units. [1][2]
Alongside the clustering upgrades, Nvidia announced software inference enhancements designed for local artificial intelligence development. The company claims these optimizations will deliver up to 40% faster performance for agentic AI workloads across select RTX hardware, with matching improvements arriving for DGX Spark and DGX Station platforms. [1][2]
Nvidia has expanded DGX Spark's clustering capabilities, allowing users to connect up to eight systems for a combined 1 TB of unified memory. The company has also announced inference optimisations promising up to 40% faster agentic AI workloads on select RTX hardware, alongside improvements for DGX Spark and DGX Station.
via Notebookcheck
NVIDIA has announced DGX Spark platforms now support 8-way clusters while AI inference gains a 40% boost on DGX & RTX systems. 8-Way NVIDIA DGX Spark Clusters Now Supported, Boosting Memory To 512 GB on 64 GB & 1 TB on 128 GB Configurations Today, NVIDIA is sharing new updates for its DGX and RTX platforms on the local AI development front.
via Wccftech