NVIDIA data center Available

NVIDIA GB200 NVL72

GB-Series ยท Blackwell Architecture

Enquire Now
The GB200 NVL72 is NVIDIA's rack-scale AI supercomputer combining 36 Grace CPUs with 72 Blackwell GPUs connected via NVLink 5.0. This system delivers 1 exaflop of AI inference performance and 13.5TB of fast GPU memory, purpose-built for training and deploying the world's largest AI models.

Key Features

Grace CPU + 2x Blackwell GPUs 72 GPU rack-scale system NVLink 5.0 fabric 13.5 TB aggregate HBM3e 1 Exaflop AI inference per rack

Full Specifications

Compute

Architecture Blackwell + Grace
Process Node 4nm TSMC
CUDA Cores per GPU 18,432
Tensor Cores 576
FP32 Performance 90 TFLOPS
FP16 Performance 1800 TFLOPS
BF16 Performance 1800 TFLOPS
INT8 Performance 3600 TOPS

Memory

Memory Size 192 GB
Memory Type HBM3e (per GPU)
Memory Bandwidth 8000 GB/s

Power & Physical

TDP 2700W
Form Factor Grace Blackwell Superchip
Power Connectors Custom NVLink

Features & Connectivity

NVLink Support Yes
Multi-GPU Support Yes

Availability

MSRP (USD) Contact for pricing
Release Date Jan 2025
Status Available

Industries

AI/ML Cloud Computing Research

Use Cases

Trillion-parameter Models Sovereign AI Infrastructure AI Factory

Interested in the NVIDIA GB200 NVL72?

Get pricing, availability, and bulk discount information from our team.

Enquire Now

Related GPUs

NVIDIA data center

NVIDIA H100 SXM

Memory

80GB HBM3

FP32

66.91 TFLOPS

TDP

700W

FP16

989.4 TFLOPS

Available View Specs
NVIDIA data center

NVIDIA H100 PCIe

Memory

80GB HBM3

FP32

51.22 TFLOPS

TDP

350W

FP16

756 TFLOPS

Available View Specs
NVIDIA data center

NVIDIA H200 SXM

Memory

141GB HBM3e

FP32

66.91 TFLOPS

TDP

700W

FP16

989.4 TFLOPS

Available View Specs
NVIDIA data center

NVIDIA B200

Memory

192GB HBM3e

FP32

90 TFLOPS

TDP

1000W

FP16

1800 TFLOPS

Available View Specs