S Sardaana
← Back to Sardaana AI
AI News

Nvidia adds 64GB DGX Spark GB10 config from ~$4,999 as RAM prices climb

On 2 October 2026 Nvidia announced a 64GB unified-memory DGX Spark configuration using the same GB10 Grace Blackwell Superchip as the 128GB model; partner systems from Acer, ASUS, Dell, Gigabyte, HP and MSI start at $4,999 on 23 October.

On 2 October 2026, Nvidia said a new 64GB unified-memory configuration of DGX Spark — built around the same GB10 Grace Blackwell Superchip, DGX OS and NVIDIA AI software stack as the 128GB model — will ship from manufacturer partners starting Friday, 23 October, at prices beginning at $4,999.

What stays the same

According to Nvidia’s blog, the 64GB SKU keeps ConnectX-7 networking, the full CUDA-accelerated software stack, and support for local agents, inference and fine-tuning. Nvidia says a single 64GB unit can run models up to about 100 billion parameters on device. Tom’s Hardware notes the same 20-core Arm CPU complex and about 273 GB/s shared memory bandwidth as the original Spark, and that skyrocketing RAM prices have made pairing the SoC with 128GB of costly LPDDR5X a steeper barrier for local AI builders who do not need that much memory.

Partners, clustering and price context

Systems will come from Acer, ASUS, Dell, Gigabyte, HP and MSI. Nvidia’s Sync Cluster Assistant can link two 64GB units over ConnectX-7 to pool memory to 128GB (and, in Nvidia’s Qwen 3.8 27B test, up to about 1.7× performance versus one unit). Tom’s Hardware reports that in-stock 128GB GB10 systems can run roughly $7,000–$9,000; HotHardware says Nvidia’s own 128GB list has moved toward about $6,995. The 128GB Spark remains available for larger models and fine-tuning. This article summarises company and trade press announcements — not a product endorsement.

Also available in: Саха тылаРусскийEspañol中文Portuguêsالعربية

Sign in to Sardaana

An account is only needed to publish, reply, vote and collaborate.

You can browse the site and read news, projects and the forum without registering.