Hardware Datasheet · GPU Server
NVIDIA
NVIDIA DGX H200
Eight H200 GPUs with 1,128GB of HBM3e — the memory-heavy Hopper DGX.
Overview
DGX H200 keeps the Hopper-generation DGX chassis and operational model while raising GPU memory substantially: 141GB of HBM3e per GPU for 1,128GB across the node, against 640GB on DGX H100. For inference on large models that difference frequently removes GPUs from every node, which is where the economics turn. We supply it configured, with fabric and storage quoted alongside.
Specifications
| GPU / Accelerator | 8× NVIDIA H200 SXM |
| GPU Memory | 1,128GB HBM3e total (141GB per GPU) |
| GPU Interconnect | NVLink with NVSwitch |
| CPU | Dual Intel Xeon Platinum |
| System Memory | 2TB DDR5 (configuration confirmed at quote) |
| Networking | ConnectX-7 400Gb/s InfiniBand; BlueField-3 DPUs |
| Form Factor | 8U rackmount |
| Sourcing | Configured and supplied by Nexus Compute; availability confirmed with quote |
Typical Use Cases
- ·Large-model inference serving with long context windows
- ·Memory-bound training and fine-tuning
- ·Extending an existing DGX H100 cluster
- ·Consolidating inference capacity onto fewer nodes
Industries
Warranty & Support
Supplied through authorized channels with full manufacturer warranty. On-site, next-business-day support options available. Every system is configured, tested, and documented before delivery, with asset and warranty records provided for enterprise audit requirements.
Request a tailored quote
Configurations are tailored per engagement — contact us for pricing and lead times.
sales@nexus-compute.com
+1 737 276 1016
nexus-compute.com
Specifications are indicative and configured to each engagement. All product names, logos, and trademarks are the property of their respective owners. Nexus Compute is an independent enterprise hardware supplier.