Nvidia is introducing a cut down version of the DGX Spark with half the memory and storage to provide a more affordable AI system. The GB10 based systems will be offered exclusively through hardware partners including Acer, Asus, Dell, Gigabyte, HP, and MSI, and are expected to retail for around $4,999.
The less powerful SKUs are necessary after Nvidia jacked the price of its 128 GB DGX Spark on Friday to $6,950, which is an increase of nearly 75 percent from last year. Skyrocketing memory prices are to blame for the massive price adjustment. That makes the 64 GB model considerably less than Nvidia's top spec DGX Spark and GB10 systems given the ongoing memory shortage, but that is still 25 percent more than the 128 GB version retailed for at launch.
Nvidia's GB10 platform has faced pricing creep since the Projects Digits concept was unveiled at CES last year. Originally, the appliance was expected to retail for around $3,000 rather than the $4,000 it eventually ended up selling for. Even at that price, the DGX Spark offered large quantities of GPU memory, making it technically the highest capacity workstation GPU Nvidia sold at launch.
Read nextNvidia Share Price Reaches Record High and Pushes Market Value to $5.7 TrillionThe smaller model supports local AI inference
The 64 GB model features a smaller capacity and is not as well suited to certain AI workloads like fine tuning. However, for local AI inference powering private agents, Nvidia argues models in the 26 to 35 billion parameter range, such as Qwen 3.8 27B, are now good enough to get the job done. The memory bandwidth remains unchanged at 273 GB/s, indicating the use of lower capacity LPDDR5x memory modules rather than fewer of them.
Beyond the new memory configuration, the systems are otherwise unchanged and feature the same 20 core Arm processor from MediaTek. They also retain onboard ConnectX 7 networking, enabling clustering of up to 4x GB10 based devices at 200 Gbps a piece. Nvidia states it will roll out new software tools to DGX OS later this month to facilitate cluster configurations and inference deployments without requiring CLI tinkering.
Upcoming notebooks could face higher prices
The lower MSRP of the 64 GB version makes systems more accessible amid the ongoing memory crunch, though it does not bode well for Nvidia's upcoming RTX Spark notebooks and mini PCs. These systems use the same GB10 silicon for general purpose computing like AAA gaming, web browsing, and content creation. Announced at Computex late this spring, RTX Spark systems were expected to be expensive, and DGX Spark pricing nearing $7,000 suggests similarly equipped notebooks could sell for significantly more.
Nvidia will offer versions of the chip with much lower memory capacities for price conscious PC buyers across the DGX and RTX Spark lines. Both lines also face competition from AMD's newly launched Gorgon Halo SoCs, which offer memory capacities ranging from around 32 GB at the low end to 192 GB for top spec Ryzen AI Max+ 495 SKUs. Systems powered by that chip currently retail for less than Nvidia's 128 GB DGX Spark, though testing shows the GB10 GPU offers substantially higher performance for AI tasks.
