NVIDIA
NVIDIA HGX A100 8-GPU (Supermicro AS-4124GO-NART)
NVIDIA HGX A100 8-GPU server — 640 GB HBM2e per node on the Supermicro AS-4124GO-NART, the deep secondary-market workhorse for training and inference.
NVIDIA HGX A100 8-GPU (Ampere) server on the Supermicro AS-4124GO-NART 4U platform. Eight A100 SXM4 GPUs deliver 80 GB of HBM2e each — 640 GB per node — fully connected over a 3rd-generation NVLink baseboard for training and inference. Dual AMD EPYC 7003 (Milan) processors host the accelerators. A deep secondary-market platform as hyperscalers rotate to Hopper and Blackwell; unit grade and hours are confirmed per node at quote. Configuration and availability confirmed at quote.
Specifications
| GPU architecture | NVIDIA A100 (Ampere) |
|---|---|
| GPUs per node | 8× NVIDIA HGX A100 80 GB SXM4 |
| GPU memory | 80 GB HBM2e per GPU · 640 GB per node · ~2.0 TB/s per GPU |
| CPU | 2× AMD EPYC 7003 (Milan) |
| NVLink fabric | 3rd-gen NVLink baseboard · 600 GB/s per GPU |
| Scale-out networking | Up to 8× 200 Gb/s HDR InfiniBand (configurable) |
| Form factor | 4U rackmount, air-cooled · unit of sale: 8-GPU node |
| Cooling | Air-cooled |
| Max system power | Up to ~6 kW (3,000 W ×2, 2+2 redundant PSUs) |
| System weight | ≈75 kg (166 lb) net |
| Warranty | Warranty scope confirmed per unit at quote |
Specifications shown are headline figures. Confirm full configuration details when you request a quote.
Last updated
Request a quote for NVIDIA HGX A100 8-GPU (Supermicro AS-4124GO-NART)
Tell us your requirements and we'll get back to you with availability and pricing.
or reach us directly at