Since its release, NVIDIA has continued to push performance of the Grace Blackwell-powered DGX Spark through continuous software optimization and close collaboration with software partners and the open-source community. These efforts are delivering meaningful gains across inference, training and creative workflows.
At CES 2026, the latest DGX Spark software release, combined with new model updates, and open-source libraries provide significant performance improvements for both DGX Spark and OEM GB10-based systems.
Scaling large models locally with unified memory and NVFP4
DGX Spark is designed for working locally with large models, featuring 128GB of unified memory in a compact desktop form factor. Two DGX Spark systems can be connected to deliver 256GB of combined memory, enabling developers to run even larger models locally.
The systems connect using ConnectX-7 networking, providing 200 Gbps of bandwidth for fast, low-latency multi-node workloads.
