NVIDIA GB200 NVL72

The NVIDIA GB200 NVL72, built on the new Blackwell architecture paired with Arm-based Grace CPU, is capable of delivering delivering 30X faster real-time
large language model (LLM) inference, 25X lower TCO, and consumes 25X less energy.
We will include pricing when this product will be available to the public.

Category:

Description

// solution overview

NVIDIA GB200 NVL72 System

The NVIDIA GB200 NVL72 is a perfect solution for large-scale AI applications as it can handle even trillion-parameter datasets with ease. Based on the new Blackwell architecture paired with Arm Grace CPU, the final superchip is capable of groundbreaking performance while keeping the TCO and power consumption up to 25x lower compared to previous generation systems.

performance
configurability
TCO
premium support
// Key features

Substantial performance gain with GB200 NVL72

LLM Inference

vs H100 Tensor Core GPU*
0 X

LLM Training

vs H100*
0 X

Energy Efficiency

vs H100*
0 X

Data Processing

vs CPU*
0 X

*NVIDIA: LLM inference and energy efficiency: TTL = 50 milliseconds (ms) real time, FTL = 5s, 32,768 input/1,024 output, NVIDIA HGX™ H100 scaled over InfiniBand (IB) vs. GB200 NVL72, training 1.8T MOE 4096x HGX H100 scaled over IB vs. 456x GB200 NVL72 scaled over IB. Cluster size: 32,768
A database join and aggregation workload with Snappy / Deflate compression derived from TPC-H Q4 query. Custom query implementations for x86, H100 single GPU and single GPU from GB200 NLV72 vs. Intel Xeon 8480+
Projected performance subject to change.

// Hardware

NVIDIA GB200 NVL72 Configurations

GB200 NVL72GB200 Superchip
Configuration36x Grace CPU, 72x B200 GPU1x Grace CPU, 2x B200 GPU
FP4 Tensor Core*1,440 PFLOPS40 PFLOPS
FP8 / FP6 Tensor Core*720 PFLOPS20 PFLOPS
INT8 Tensor Core*720 POPS20 POPS
FP16 / BF16 Tensor Core*360 PFLOPS10 PFLOPS
TF32 Tensor Core*180 PFLOPS5 PFLOPS
FP64 Tensor Core3,240 TFLOPS90 TFLOPS
GPU MemoryUp to 13.5 TB HBM3e, 576 TBpsUp to 384 GB HBM3e, 16 TBps
NVLink Bandwidth130 TBps3.6 TBps
CPU Cores2,952 Arm Neoverse V2 Cores72 Arm Neoverse V2 Cores
CPU MemoryUp to 17 TB LPDDR5X, Up to 18.4 TBpsUp to 480 GB LPDDR5X, Up to 18.4 TB/s
Product infoDatasheet
// Networking

High speed infiniband

GB200 NVL72 supports advanced networking options at speeds up to 800 gigabits per second (Gb/s). For the highest AI performance, GB200 supports the latest NVIDIA Quantum-X800 InfiniBand and Spectrum™-X800 Ethernet platforms. GB200 NVL72 also includes NVIDIA BlueField-3 data processing units (DPUs) to enable cloud networking, composable storage, zero-trust security and GPU compute elasticity in hyperscale AI clouds.

NVIDIA Quantum X800 Infiniband switch

Specification

CPU

NVIDIA Grace CPU, 72 Arm Neoverse V2 cores @ 3.1 GHz

RAM

480 GB LPDDR5X

NVMe Capacity

TB/A

GPU

2x NVIDIA B200 192 GB HBM3e

Network Card

NVIDIA Bluefield-3 DPU

Reviews

There are no reviews yet.

Be the first to review “NVIDIA GB200 NVL72”

Your email address will not be published. Required fields are marked *