Supermicro AS-4125GS-TNRT 4x H100 NVL PCIe (Dual EPYC)
Flexible 4U PCIe GPU server on dual EPYC for inference and mixed AI.
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.
4x NVIDIA H100 NVL 94GB PCIe (Gen5)
Dual AMD EPYC 9004/9005, up to 128 cores per socket
Up to 6TB DDR5-6000 ECC RDIMM (24 DIMM)
24x 2.5" hot-swap NVMe/SATA/SAS bays
Overview
This 4U Supermicro A+ platform runs four H100 NVL PCIe GPUs on dual AMD EPYC 9004 CPUs, giving teams a flexible, serviceable node for inference and mixed AI without the cost of an SXM5 baseboard. Nexus Compute configures the GPU, memory, and fabric to your workload, validates the build, and warranty-backs it through authorized Supermicro channels.
Who This Solution Is For
Business Benefits
Cost-efficient acceleration
PCIe H100 NVL cards deliver strong inference and fine-tuning value without SXM5-class infrastructure overhead.
Configuration flexibility
Open PCIe Gen5 slots let us tailor GPU count and networking to your exact workload mix.
EPYC core density
Dual AMD EPYC 9004 CPUs supply abundant cores and PCIe lanes for data pipelines feeding the GPUs.
Typical Business Use Cases
Production LLM and generative inference
Model fine-tuning at moderate scale
Mixed AI, analytics, and HPC workloads
Departmental shared GPU compute
Industry Applications
Technical Overview
The AS-4125GS-TNRT is a 4U dual-socket AMD EPYC 9004/9005 platform with PCIe Gen5 x16 direct-attached GPU slots and optional NVLink bridges across H100 NVL pairs. It supports up to 6TB of DDR5-6000 ECC across 24 DIMMs, 24 hot-swap NVMe/SATA bays, and four Titanium 2000W PSUs.
| GPU / Accelerator | 4x NVIDIA H100 NVL 94GB PCIe (Gen5) |
| GPU Interconnect | PCIe Gen5 x16 direct-attach; NVLink bridge on NVL pairs |
| CPU | Dual AMD EPYC 9004/9005, up to 128 cores per socket |
| Memory | Up to 6TB DDR5-6000 ECC RDIMM (24 DIMM) |
| Storage | 24x 2.5" hot-swap NVMe/SATA/SAS bays |
| Networking / Fabric | Dual 10GbE onboard; PCIe Gen5 slots for 100/400G NICs |
| Form Factor | 4U rackmount |
| Power | 4x 2000W Titanium redundant PSUs |
| 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 do PCIe H100 NVL cards differ from SXM5?
PCIe NVL GPUs connect over PCIe Gen5 with NVLink bridges across pairs rather than a full NVSwitch fabric; they suit inference and mid-scale fine-tuning at lower infrastructure cost.
Why EPYC instead of Xeon here?
Dual EPYC 9004/9005 offers high core counts and abundant PCIe Gen5 lanes, which keep data pipelines and multiple PCIe GPUs well fed.
Can it run fewer GPUs now and grow later?
Yes. The open PCIe layout lets us populate two or four GPUs initially and add cards or faster networking as workloads expand.
Hardware Assistance
Configure the Supermicro AS-4125GS-TNRT 4x H100 NVL PCIe (Dual EPYC) 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.