DGX Spark
AI workstation, GB10 superchip
| Memory | 128 GB unified LPDDR5X |
|---|---|
| Compute | 1 PFLOP FP4 (1000 TOPS) |
| Chip | GB10 Grace Blackwell, 20 Arm cores |
| Network | ConnectX-7 200 GbE, cluster of 1 to 4 nodes |
| Power draw | 170 W, wall outlet |
| Form factor | 150 mm desktop |
What the machine runs
Models up to 200 billion parameters quantised on a single node. Directly connecting two nodes lets the platform address up to 405 billion parameters, and a ConnectX-7 switch extends the cluster further. Recommended runtime: llama.cpp or vLLM locally.
Positioning
A sovereign workstation for a small business puts a private model on the desk, with no reliance on the cloud.
Manufacturers
NVIDIA offers the DGX Spark through its partners, including ASUS, Dell, Gigabyte, HP, Lenovo and MSI.
GPUDirect Storage
On platforms fitted with ConnectX cards, GPUDirect Storage technology establishes a direct path between NVMe or NVMe over Fabric storage and GPU memory, bypassing the CPU buffer. Sustained throughput reaches around 50 gigabytes per second per link. Compatible storage arrays, such as NetApp, VAST, DDN or WEKA, feed training and large-scale RAG at full speed.
Support
Open stack supported by QDNA and the communities. Optional NVIDIA AI Enterprise (NIM, NeMo, Triton) with a per-card service-level agreement.