Europe
Availability on request
Price on requestConfiguration and region to be confirmed
GLOBALATLAS AI · GB200 NVL72
NVIDIA GB200 NVL72 · 186 GB
GB200 NVL72 is a liquid-cooled rack with 72 Blackwell GPUs and 36 Grace CPUs. NVIDIA specifies 13.4 TB total GPU memory, about 186 GB per GPU. This is not a single GPU server.
Grace Blackwell186GB / GPU
- Hardware variant
- 72-GPU rack / 36 Grace CPUs
- Memory technology
- HBM3e
- Workloads
- Rack-scale AI training and large-model serving
Configuring GB200 NVL72
- 01
Define the complete 72-GPU NVLink domain
- 02
Confirm rack power, liquid cooling and facility compatibility
- 03
Specify the cluster scheduler, network and checkpoint storage
Regional configurations
USA
Availability on request
Price on requestConfiguration and region to be confirmed
Asia
Availability on request
Price on requestConfiguration and region to be confirmed
Technical references
Specification 1 →YOUR CONFIGURATION
Request an infrastructure quote
Choose resources and operating time. We will confirm a price for your configuration.
Each resource has its own listed configuration. Multiple GPU servers do not form an NVLink cluster. Monthly rates cover the full billing period; storage and traffic outside the configuration are not included.