Supermicro AS-4125GS-TNRT 4U with 8x NVIDIA H100 NVL
Eight H100 NVL accelerators in a flexible 4U EPYC platform
We help you choose, configure, and deliver the right system — no obligation.
Configuration at a Glance
Tailored per engagement. Full technical overview below.
Configuration Options
Core specifications for this system. Every component is configurable to your workload — request a quote for a tailored build.
8x NVIDIA H100 NVL (94GB HBM3 each)
2x AMD EPYC 9005/9004, up to 160C/320T
Up to 6TB DDR5 ECC RDIMM (24 DIMM slots)
4x 2.5" NVMe + 2x 2.5" SATA hot-swap, 1x M.2
Overview
The AS-4125GS-TNRT is a 4U dual-socket AMD EPYC GPU server configured with eight double-width NVIDIA H100 NVL accelerators on direct-attached PCIe 5.0 x16 lanes. Each unit is specified, assembled, burned-in, and validated against the target workload, then delivered with full manufacturer warranty through authorized distribution channels.
Who This Solution Is For
Business Benefits
Validated on arrival
Every node is thermally stress-tested and firmware-aligned before it leaves the integration bench.
Flexible accelerator bay
The direct-attach PCIe topology lets you mix H100 NVL, L40S, or RTX PRO 6000 cards as roadmaps shift.
Channel-backed warranty
Hardware ships with full Supermicro warranty and documented authorized-source provenance for audit.
Typical Business Use Cases
LLM fine-tuning and inference serving
Multi-GPU deep learning training
Scientific HPC simulation
Retrieval-augmented generation pipelines
Industry Applications
Technical Overview
Built on the Supermicro H13 dual-socket AMD EPYC 9005/9004 platform, the system direct-attaches eight double-width GPUs over PCIe 5.0 x16 with optional NVLink Bridge pairing for adjacent cards. DDR5 RDIMM across 24 slots and front NVMe bays feed the accelerators while four redundant Titanium power supplies sustain a full GPU load.
| GPU/Accelerator | 8x NVIDIA H100 NVL (94GB HBM3 each) |
| GPU Interconnect | PCIe 5.0 x16 direct-attach, optional NVLink Bridge |
| CPU | 2x AMD EPYC 9005/9004, up to 160C/320T |
| Memory | Up to 6TB DDR5 ECC RDIMM (24 DIMM slots) |
| Storage | 4x 2.5" NVMe + 2x 2.5" SATA hot-swap, 1x M.2 |
| Networking | 2x 10GBASE-T, dedicated 1GbE BMC port |
| Form Factor | 4U rackmount, air-cooled |
| Power | 4x 2000W redundant Titanium (2+2 / 3+1) |
| Warranty | Manufacturer warranty via authorized channel |
Specifications are indicative and configured to each engagement. Request a quote for a configuration tailored to your requirements.
Warranty, Support & Fulfillment
Every system ships from an authorized channel, configured and tested, with the documentation enterprise buyers need — backed by warranty and a dedicated account team.
Enterprise Warranty
Full manufacturer warranty with optional on-site, next-business-day support and extended coverage.
Authorized Channel
Sourced through Tier-1 distribution and OEM partners — never grey market. Asset & warranty records included.
Lead Time & Deployment
48-hour quotes, then configured, burn-in tested, and delivered on a committed schedule.
Nationwide Fulfillment
Coordinated logistics, rack-and-stack, and delivery wherever your infrastructure lives.
Frequently Asked Questions
How many GPUs does the AS-4125GS-TNRT support?
It supports up to eight double-width PCIe GPUs on direct-attached PCIe 5.0 x16 lanes; we configure all eight H100 NVL cards in this build.
Is this better suited to training or inference?
The eight H100 NVL accelerators handle both demanding multi-GPU training jobs and high-throughput LLM inference, making it a versatile single-node AI workhorse.
Can it be deployed in a standard air-cooled rack?
Yes, the 4U chassis is air-cooled and rated for typical data-center ambient conditions, so no liquid-cooling infrastructure is required.
Hardware Assistance
Configure the Supermicro AS-4125GS-TNRT 4U with 8x NVIDIA H100 NVL with Nexus Compute
Tell us your requirements and a hardware specialist will help you specify, configure, and quote the right system — typically within two business days. No obligation.