NVIDIA
NVIDIA HGX H100 8-GPU Liquid-Cooled Server (Supermicro 4U H13)
NVIDIA HGX H100 8-GPU server — liquid-cooled on the Supermicro 4U H13 with 8× H100 80GB SXM5, AMD EPYC Genoa, and ConnectX-7 NDR400 networking.
Condition: New. Lead time: 2–4 weeks.Request a quote →
Liquid-cooled 8-GPU NVIDIA HGX H100 server, on the Supermicro 4U H13 with a DELTA-NEXT Vulcan HGX baseboard and per-GPU cold plates. Eight H100 80GB SXM5 GPUs give 640 GB HBM3 per node, paired with dual AMD EPYC Genoa CPUs and 8× ConnectX-7 NDR400 networking for dense, efficient AI compute. Configuration and availability confirmed at quote.
Specifications
| GPU architecture | NVIDIA H100 (Hopper) |
|---|---|
| GPUs per node | 8× H100 80 GB SXM5 |
| GPU memory | 80 GB HBM3 per GPU · 640 GB per node |
| CPU | 2× AMD EPYC Genoa 9454 · 48C/96T · 2.75 GHz · 290 W |
| NVLink fabric | 4th-gen NVLink baseboard · 900 GB/s per GPU |
| Scale-out networking | 8× ConnectX-7 NDR400 OSFP · 3.2 Tb/s per node · InfiniBand / Spectrum-X |
| Form factor | 4U rackmount, liquid-cooled · unit of sale: 8-GPU node |
| Cooling | Direct liquid cooling · DELTA-NEXT Vulcan baseboard · per-GPU cold plates |
| GPU board power | 8× 700 W H100 SXM5 (5.6 kW GPU baseboard) |
| Condition | New |
| Warranty | 3-yr onsite Next Business Day · 3-yr NVIDIA Enterprise (per ConnectX-7) |
Specifications shown are headline figures. Confirm full configuration details when you request a quote.
Last updated
Request a quote for NVIDIA HGX H100 80GB — Liquid-Cooled (Supermicro 4U H13)
Tell us your requirements and we'll get back to you with availability and pricing.
or reach us directly at