The NVIDIA DGX GB200 is a rack-scale AI factory purpose-built for training and inferencing trillion-parameter generative AI models. It integrates 36 NVIDIA GB200 Grace Blackwell Superchips into a single liquid-cooled system, providing massive supercomputing performance for mission-critical enterprise AI workloads.
- Features 72 NVIDIA Blackwell GPUs and 36 NVIDIA Grace CPUs with 2,592 Arm Neoverse V2 cores.
- Delivers up to 1,440 PFLOPS of FP4 Tensor Core performance for state-of-the-art AI models.
- Provides up to 13.4 TB of HBM3e GPU memory with a massive 576 TB/s aggregate bandwidth.
- Utilizes fifth-generation NVLink to achieve 1.8 TB/s of GPU-to-GPU bidirectional bandwidth.
- Includes 72x OSFP single-port NVIDIA ConnectX-7 VPI adapters supporting 400 Gb/s InfiniBand.
- Equipped with 36x dual-port NVIDIA BlueField-3 VPI adapters for 200 Gb/s InfiniBand and Ethernet.
- Features 9x L1 NVIDIA NVLink Switches for high-speed, low-latency interconnectivity.
- Managed via NVIDIA Mission Control for streamlined AI factory operations and resilience.
- Optimized for liquid cooling to ensure maximum efficiency in high-density deployments.
- Fully supported by NVIDIA AI Enterprise and DGX OS for a production-ready software stack.
NVIDIA - DGX GB200 - Rack-Scale AI Infrastructure System
For a quote, contact us at info@tropical.com
- SKU:
- DGX GB200
- Weight:
- 2,850.00 LBS
Related Products
A relatable product
NVIDIA - DGX SuperPOD (GB200) - AI Infrastructure Supercomputer
The NVIDIA DGX SuperPOD with GB200 systems is a leadership-class, liquid-cooled AI infrastructure designed for the most challenging generative AI training and inference workloads. It provides a...
NVIDIA - DGX BasePOD (GB200) - Blackwell AI Infrastructure Foundation
NVIDIA DGX BasePOD is a validated reference architecture designed to simplify the deployment and scaling of enterprise AI infrastructure. Built with NVIDIA Blackwell GB200 systems, it provides a...
NVIDIA - DGX GB200 NVL72 - Rack-Scale Liquid-Cooled AI Supercomputer
The NVIDIA GB200 NVL72 is a rack-scale, liquid-cooled exascale computer designed for real-time trillion-parameter large language model (LLM) inference and massive-scale AI training. It integrates 36...
NVIDIA - DGX A100 - AI Infrastructure System
The NVIDIA DGX A100 is the universal system for all AI workloads, offering unprecedented compute density, performance, and flexibility in the world's first 5-petaFLOPS AI system. It integrates eight...
NVIDIA - DGX H200 - AI Infrastructure System
The NVIDIA DGX H200 is the premier AI factory infrastructure designed for enterprise-scale generative AI, natural language processing, and deep learning. It integrates eight NVIDIA H200 Tensor Core...