GB200 NVL72
GB200 NVL72 connects 36 Grace CPUs and 72 Blackwell GPUs in a rack-scale design. The GB200 NVL72 is a liquid-cooled, rack-scale solution that boasts a 72-GPU NVLink domain that acts as a single massive GPU and delivers 30X faster real-time trillion-parameter LLM inference.
GB200 NVL72
Infrastructure and technology partners
Perfect for a range of workloads
Deploying AI based workloads on CUDO Compute is easy and cost-effective. Follow our AI related tutorials.
Deploying rendering based workloads on CUDO Compute is easy and cost-effective.
From video editing to image generation, virtualization is ideal for your content creation needs.
Purpose-built GB200 clusters, designed and managed by CUDO
Deployed across 16 ISO-certified data centres
From 8 to 1,000+ GPUs in a single deployment
NVIDIA Quantum-X800 InfiniBand or Spectrum-X Ethernet networking
Expert rack-level design, installation, and benchmarking before handoff
24/7 monitoring, management, and engineering support
Compatible with Slurm, Kubernetes, and NVIDIA Base Command
Available at the most cost-effective pricing
Launch your AI products faster with on-demand GPUs and a global network of data center partners
Bare metal
Powered by renewable energy
No noisy neighbors
SpectrumX local networking
300Gbps external connectivity
NVMe SSD storage
Enterprise
Powerful GPU clusters
Scalable data center colocation
Large quantities of GPUs and hardware
Optimize to your requirements
Expert installation
Scale as your demand grows
Specifications
Browse specifications for the NVIDIA GB200 GPU
Starting from
Contact us for pricing
Architecture
NVIDIA Blackwell
GPU
72x NVIDIA Blackwell GPUs
GPU memory
13.5 TB total HBM3e, 576 TB/s aggregate bandwidth
FP4 tensor core performance
1,440 PFLOPS
FP8 tensor core performance
720 PFLOPS
NVIDIA NVSwitch
9x L1 NVIDIA NVLink Switches (Fifth-generation)
NVIDIA NVLink bandwidth
130 TB/s aggregate intra-rack bandwidth (1.8 TB/s per GPU, allowing all 72 GPUs to communicate as one)
System power usage
~120 kW rack TDP
CPU
36x NVIDIA Grace CPUs (2,592 Arm Neoverse V2 cores total)
System memory
Up to 17 TB LPDDR5X, up to 18.4 TB/s aggregate bandwidth
Networking
Up to 72x OSFP single-port NVIDIA ConnectX-7 or ConnectX-8 VPI (400 Gb/s to 800 Gb/s NVIDIA InfiniBand/Ethernet) Up to 18x dual-port NVIDIA BlueField-3 VPI DPUs
Management network
Host baseboard management controller (BMC) with RJ45 per tray, 2x Top-of-Rack (TOR) out-of-band management switches
Storage
Per compute tray (18 compute trays total): 8x E1.S Gen5 NVMe drive bays for internal storage, 1x M.2 NVMe SSD for OS boot
Software
NVIDIA AI Enterprise (optimized AI software), NVIDIA Mission Control (AI data center operations/orchestration), NVIDIA Base Command, Ubuntu / RHEL
Rack units (RU)
48
Operating temperature
Requires dedicated Direct-to-Chip (D2C) liquid cooling (CPUs, GPUs, and NVSwitches are liquid-cooled via an in-rack or in-row Coolant Distribution Unit, while networking modules and storage are air-cooled).
Ideal uses cases for the NVIDIA GB200 NVL72 GPU
Explore uses cases for the NVIDIA GB200 NVL72 including Supercharging next-generation AI and accelerated computing, Energy-efficient infrastructure, Massive-scale training.
Supercharging next-generation AI and accelerated computing
GB200 NVL72 introduces cutting-edge capabilities and a second-generation Transformer Engine which enables FP4 AI and when coupled with fifth-generation NVIDIA NVLink, delivers 30X faster real-time LLM inference performance for trillion-parameter language models.
Energy-efficient infrastructure
Liquid-cooled GB200 NVL72 racks reduce a data center’s carbon footprint and energy consumption. Liquid cooling increases compute density, reduces the amount of floor space used, and facilitates high-bandwidth, low-latency GPU communication with large NVLink domain architectures.
Massive-scale training
GB200 NVL72 includes a faster second-generation Transformer Engine featuring FP8 precision, enabling a remarkable 4X faster training for large language models at scale.
Blog
Resources
- Emmanuel Ohiri
Resources
- Emmanuel Ohiri
Resources
- Emmanuel Ohiri
Resources
- Emmanuel Ohiri
Resources
- Emmanuel Ohiri
Resources
- Emmanuel Ohiri
Browse alternative GPU solutions for your workloads
Access a wide range of performant NVIDIA and AMD GPUs to accelerate your AI, ML & HPC workloads
NVIDIA H100 PCIe
Price on request
Scale with high performance H100 GPUs on our reserved cloud.
