NVIDIA DGX H100

rack
8× H100
NVSwitch + NVLink
2022
FP64
0.3
PFLOPS
FP32
0.5
PFLOPS
Power
8.5
kW Total
Memory
640
GB Total

Enterprise AI system with 8x H100 SXM5 GPUs

We will point you at suppliers who have it. Free, and no signup.

FP64
0.27
PFLOPS
32.0 TFLOPS/kW
FP32
0.54
PFLOPS
63.1 TFLOPS/kW
TF32
3.96
PFLOPS
930.8 TFLOPS/kW
FP16
7.92
PFLOPS
1862.6 TFLOPS/kW

System Details

GPU Configuration

GPU Count: 8 GPUs
Architecture: Hopper
Interconnect: NVSwitch + NVLink

System Specifications

Form Factor: rack
Total Power: 8.5 kW
Total Memory: 640 GB
Memory Bandwidth: 26.8 TB/s

Precision Performance Breakdown

PrecisionSystem PerformancePer GPUEfficiency
0.272 PFLOPS 34.0 TFLOPS 32.0 TFLOPS/kW
0.536 PFLOPS 67.0 TFLOPS 63.1 TFLOPS/kW
3.956 PFLOPS 494.5 TFLOPS 930.8 TFLOPS/kW
7.916 PFLOPS 989.5 TFLOPS 1862.6 TFLOPS/kW
7.916 PFLOPS 989.5 TFLOPS 1862.6 TFLOPS/kW
15.832 PFLOPS 1979.0 TFLOPS 3725.2 TFLOPS/kW
15.832 PFLOPS 1979.0 TFLOPS 3725.2 TFLOPS/kW

Where a vendor states a basis, the figure shown is the dense one, and any figure it publishes is listed separately. Vendors commonly headline the sparse number instead: for NVIDIA's 2:4 structured sparsity that is exactly twice the dense figure, though other vendors' sparse modes do not all follow that ratio.

Powered by NVIDIA H100

This system utilizes 8 × NVIDIA H100 SXM5 80GB GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

700W

Per GPU Memory

80 GB

Process Node

4nm

Architecture

Hopper

Documentation & Resources

Flopper Spec Sheet

NVIDIA DGX H100 specifications, generated from our database. Printable.

Vendor product page

NVIDIA DGX H100 documentation

View ↗

workstation-datasheet-dgx-spark-gtc25-spring-nvidia-us-3716899-web.pdf

NVIDIA • 2025-10-16

View Document ↗

NVIDIA DGX Documentation

System guides, deployment resources

Browse ↗

Typical Use Cases

Large Language Model Training
Distributed Deep Learning
Multi-GPU Inference
HPC Simulations
Scientific Computing

The NVIDIA DGX H100 runs 8× H100 GPUs over NVSwitch + NVLink, delivering 15.8 PFLOPS FP8 dense.