What We Know
Nvidia is introducing a DGX Spark configuration with 64GB of unified memory to give developers another option for building and scaling AI workloads locally.3Backed by 3 sourcesblogs.nvidia.commediarelease.cotoolboxkart.tech The 64GB system is positioned as a lower-capacity alternative to the 128GB DGX Spark, while retaining the same underlying platform according to reporting about the new configuration.2Backed by 2 sourcestoolboxkart.techtechgenyz.com
The new systems are reported to start at $4,999 and to become available through Acer, ASUS, Dell, Gigabyte, HP, and MSI on October 23.2Backed by 2 sourcestechgenyz.comaicoder.com The expanded lineup is relevant to developers comparing local AI hardware with alternatives such as Mac Studio systems, Strix Halo PCs, or multiple DGX Spark units.2Backed by 2 sourcesmindstudio.aicomputingforgeeks.com
Source Comparison
Aligned reportingCorroborates
- blogs.nvidia.com↗The Nvidia announcement directly supports the introduction of a 64GB DGX Spark option for building and scaling local AI workloads.
- mediarelease.co↗The release reproduces the announcement that Nvidia is offering a 64GB unified-memory DGX Spark configuration for local AI development.
- mindstudio.ai↗The comparison directly supports the relevance of DGX Spark alongside other local AI hardware options for developers evaluating local language-model systems.
- computingforgeeks.com↗The article explicitly frames DGX Spark, Strix Halo, and Mac Studio as alternatives in a local-AI hardware comparison.
- toolboxkart.tech↗The article supports both the 64GB local-AI configuration and its positioning as an alternative to the 128GB system.
- techgenyz.com↗The report gives the stated starting price, launch date, participating manufacturers, and the same-platform relationship to the 128GB model.
- aicoder.com↗The report independently repeats the $4,999 price and October 23 OEM availability for the 64GB DGX Spark.