NVIDIA GB200 NVL32

rack
32× B200
NVLink
2025
FP64
1.3
PFLOPS
FP32
2.6
PFLOPS
Power
53.3
kW Total
Memory
GB Total

16 GB200 superchips (32 B200 GPUs + 16 Grace CPUs) in a rack-scale system

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

FP64
1.28
PFLOPS
27.0 TFLOPS/kW
FP32
2.56
PFLOPS
54.0 TFLOPS/kW
TF32
40.00
PFLOPS
1501.0 TFLOPS/kW
FP16
80.00
PFLOPS
2701.0 TFLOPS/kW

System Details

GPU Configuration

GPU Count: 32 GPUs
Architecture: Blackwell
Interconnect: NVLink

System Specifications

Form Factor: rack
Total Power: 53.3 kW
Total Memory:
Memory Bandwidth: 246.4 TB/s

Precision Performance Breakdown

PrecisionSystem PerformancePer GPUEfficiency
1.280 PFLOPS 40.0 TFLOPS 27.0 TFLOPS/kW
2.560 PFLOPS 80.0 TFLOPS 54.0 TFLOPS/kW
40.000 PFLOPS 1250.0 TFLOPS 1501.0 TFLOPS/kW
80.000 PFLOPS 2500.0 TFLOPS 2701.0 TFLOPS/kW
80.000 PFLOPS 2500.0 TFLOPS 2701.0 TFLOPS/kW
160.000 PFLOPS 5000.0 TFLOPS 5403.0 TFLOPS/kW
320.000 PFLOPS 10000.0 TFLOPS 10806.0 TFLOPS/kW
160.000 PFLOPS 5000.0 TFLOPS 5403.0 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 B200

This system utilizes 32 × NVIDIA B200 SXM 180GB GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

1000W

Per GPU Memory

180 GB

Process Node

4nm

Architecture

Blackwell

Documentation & Resources

Flopper Spec Sheet

NVIDIA GB200 NVL32 specifications, generated from our database. Printable.

Vendor product page

NVIDIA GB200 NVL32 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

AI/ML Training
High-Performance Computing
Data Analytics

The NVIDIA GB200 NVL32 runs 32× B200 GPUs over NVLink, delivering 160 PFLOPS FP8 dense.