NVIDIA
Lenovo ThinkSystem SR680a V4 — HGX B300 8-GPU Server
NVIDIA HGX B300 8-GPU Blackwell Ultra server — 2,304 GB HBM3e per node on the Lenovo ThinkSystem SR680a V4 (8U) with ConnectX-8 800 Gb/s scale-out networking.
8-GPU NVIDIA HGX B300 (Blackwell Ultra) server, built on the Lenovo ThinkSystem SR680a V4 — an 8U air/hybrid-cooled platform. Eight B300 GPUs provide 2,304 GB of HBM3e per node — 288 GB and roughly 8 TB/s per GPU — bridged by a fifth-generation NVLink baseboard at 1.8 TB/s per GPU, while NVIDIA ConnectX-8 SuperNICs deliver 800 Gb/s of scale-out networking per GPU. Dual Intel Xeon 6 processors and high-capacity DDR5 system memory feed the accelerators for large-model training and inference. Configuration and availability confirmed at quote.
Specifications
| GPU architecture | NVIDIA B300 (Blackwell Ultra) |
|---|---|
| GPUs per node | 8× NVIDIA HGX B300 |
| GPU memory | 288 GB HBM3e per GPU · 2,304 GB per node · ~8 TB/s per GPU |
| CPU | 2× Intel Xeon 6 |
| NVLink fabric | 5th-gen NVLink baseboard · 1.8 TB/s per GPU |
| Scale-out networking | NVIDIA ConnectX-8 SuperNIC · 800 Gb/s per GPU |
| Form factor | 8U rackmount, air/hybrid-cooled · unit of sale: 8-GPU node |
| Cooling | Air / hybrid-cooled |
| GPU board power | 8× ~1,100 W B300 (≈8.8 kW GPU baseboard) |
| Condition | New |
| Warranty | 3-yr Lenovo on-site hardware warranty |
Specifications shown are headline figures. Confirm full configuration details when you request a quote.
Last updated
Request a quote for NVIDIA HGX B300 (Lenovo ThinkSystem SR680a V4)
Tell us your requirements and we'll get back to you with availability and pricing.
or reach us directly at