NVIDIA

NVIDIA - GH200 - Grace Hopper Superchip AI Module

For a quote, contact us at info@tropical.com


SKU:
GH200 Enterprise AI Module
Weight:
4.41 LBS
Width:
10.50 (in)
Height:
1.60 (in)
Depth:
5.50 (in)
The NVIDIA GH200 Grace Hopper Superchip is a breakthrough accelerated processor designed for giant-scale AI and high-performance computing (HPC) applications. It integrates the NVIDIA Grace CPU and Hopper GPU architectures using the high-speed NVLink-C2C interconnect to provide a coherent memory model for demanding enterprise workloads. - Integrated NVIDIA Grace CPU and Hopper GPU architectures - 900 GB/s bidirectional NVLink-C2C coherent interface - 72 Arm Neoverse V2 CPU cores with Scalable Vector Extensions (SVE2) - Up to 624GB of total fast memory including HBM3e and LPDDR5X - Delivers up to 4 petaFLOPS of AI performance - 7X higher bandwidth than PCIe Gen5 for CPU-to-GPU communication - Supports NVIDIA AI Enterprise, HPC SDK, and Omniverse software stacks - Optimized for Large Language Models (LLMs) and Graph Neural Networks (GNNs) - Unified memory architecture simplifies programming for complex simulations - Modular design compatible with NVIDIA MGX reference architectures

Related Products

A relatable product