Europe
Availability on request
Price on requestConfiguration and region to be confirmed
GLOBALATLAS AI · H100 NVL
NVIDIA H100 NVL · 94 GB
H100 NVL has 94 GB HBM3 per GPU. The 188 GB figure describes a linked pair; it is not the memory of one card.
Hopper94GB VRAM
- Hardware variant
- NVL PCIe 94 GB per GPU
- Memory technology
- HBM3
- Workloads
- Memory-intensive inference with NVLink requirements
Configuring H100 NVL
- 01
Specify one 94 GB GPU or an NVLink-linked pair
- 02
Verify the bridge and topology in the host configuration
- 03
Plan tensor parallelism and per-device memory placement
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.