Europe
Availability on request
Price on requestConfiguration and region to be confirmed
GLOBALATLAS AI · B100
NVIDIA B100 · 192 GB
The published HGX B100 reference specifies nominal 192 GB HBM3e per GPU. Usable capacity and the exact deployed platform must be confirmed in the server specification.
Blackwell192GB VRAM
- Hardware variant
- HGX B100 nominal 192 GB
- Memory technology
- HBM3e
- Workloads
- Blackwell training and inference platforms
Configuring B100
- 01
Confirm nominal and usable memory of the supplied B100
- 02
Check Blackwell support in drivers and model runtime
- 03
Specify HGX topology, power envelope and GPU count
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.