Compare

Compare edge AI platforms

Put up to four platforms side by side. Comparisons use the same published information shown on each product page, so nothing is estimated or inferred.

1/4 selected.

Side-by-side comparison of selected edge AI platforms
AttributeNVIDIA DGX SparkNVIDIA
AI performanceUp to 1,000 TOPS inference, up to 1 PFLOP FP4 (sparse)
ProcessorNVIDIA GB10 Grace Blackwell Superchip, 20-core Arm CPU (10× Cortex-X925, 10× Cortex-A725)
GPUBlackwell GPU, 6,144 CUDA cores, 5th-gen Tensor cores
Memory128 GB LPDDR5X unified, 256-bit, 273 GB/s
Storage and networking1 TB or 4 TB NVMe M.2; 10GbE RJ-45, ConnectX-7 (2× QSFP), Wi-Fi 7
Power and size240 W external PSU (140 W SoC TDP); 150 × 150 × 50.5 mm, 1.2 kg
DeploymentDesk-side development environment or small on-premise lab
Use casesLocal large model inference, Model fine-tuning, Robotics and vision development
DetailsView product

Specifications are summarised from publicly published manufacturer material and are provided for orientation only. Always confirm current figures directly with the manufacturer before purchasing or designing in.