NVIDIA GB200 NVL72
36 GB200 superchips (72 B200 GPUs + 36 Grace CPUs) in a liquid-cooled rack-scale system
We will point you at suppliers who have it. Free, and no signup.
Systems built on this design
1 configurationNVIDIA GB200 NVL72 is the baseboard. The systems above are OEM implementations of it and share its accelerator performance; they differ in chassis, CPU, memory and cooling.
System Details
GPU Configuration
System Specifications
Precision Performance Breakdown
| Precision | System Performance | Per GPU | Efficiency |
|---|---|---|---|
| 2.880 PFLOPS | 40.0 TFLOPS | 27.0 TFLOPS/kW | |
| 5.760 PFLOPS | 80.0 TFLOPS | 54.0 TFLOPS/kW | |
| 90.000 PFLOPS | 1250.0 TFLOPS | 1500.0 TFLOPS/kW | |
| 180.000 PFLOPS | 2500.0 TFLOPS | 2700.0 TFLOPS/kW | |
| 180.000 PFLOPS | 2500.0 TFLOPS | 2700.0 TFLOPS/kW | |
| 360.000 PFLOPS | 5000.0 TFLOPS | 5400.0 TFLOPS/kW | |
| 720.000 PFLOPS | 10000.0 TFLOPS | 10800.0 TFLOPS/kW | |
| 360.000 PFLOPS | 5000.0 TFLOPS | 5400.0 TFLOPS/kW | |
| FP6 | 360.000 PFLOPS | 5000.0 TFLOPS | — |
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 72 × 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 NVL72 specifications, generated from our database. Printable.
Vendor product page
NVIDIA GB200 NVL72 documentation
workstation-datasheet-dgx-spark-gtc25-spring-nvidia-us-3716899-web.pdf
NVIDIA • 2025-10-16
NVIDIA DGX Documentation
System guides, deployment resources
Typical Use Cases
The NVIDIA GB200 NVL72 runs 72× B200 GPUs over NVLink, delivering 360 PFLOPS FP8 dense.