NVIDIA DGX Spark

Official
Grace Blackwell Desktop Workstation 2025 4nm
VRAM
128 GB
TDP
240 W
Bandwidth
273 GB/s
memory

Performance Metrics

Peak theoretical throughput by precision type

PrecisionDescriptionBitsPeak TFLOPSEfficiency
FP1616-bit floating point16100.00.417 TFLOPS/W
FP16 Efficiency
0.417 TFLOPS/W
100.0 TFLOPS / 240W

Power Specifications

TDP

240 W

Max Power

276 W

Power Connector

PCIe 8-pin

Cooling

Air

Memory Specifications

Capacity

128 GB

Type

--

Bandwidth

273 GB/s

Interface

--

Hardware & Design

Form Factor

Desktop Workstation

Architecture

Grace Blackwell

Process Node

4nm

Launch Year

2025

Variant

Standard

Market Segment

Desktop AI

Documentation & Resources

Common Use Cases

General Compute AI/ML Workloads Data Processing

The NVIDIA DGX Spark is optimized for high-performance computing tasks with Grace Blackwell architecture delivering high TFLOPS of compute power.

Where to Rent

Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.

Browse GPU Cloud Providers

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs