Skip to content
HomeSolutionsSupermicroSupermicro AS-5126GS-TNRT2 5U with 10x NVIDIA L40S
Supermicro logo
Supermicro

Supermicro AS-5126GS-TNRT2 5U with 10x NVIDIA L40S

Ten L40S GPUs for dense inference and visual computing

Full manufacturer warrantyAuthorized channel48-hour quote

We help you choose, configure, and deliver the right system — no obligation.

Supermicro AS-5126GS-TNRT2 5U with 10x NVIDIA L40S — Supermicro enterprise hardware
Supermicro logo
Supermicro AS-5126GS-TNRT2 5U with 10x NVIDIA L40S hardware detail 1
Supermicro AS-5126GS-TNRT2 5U with 10x NVIDIA L40S hardware detail 2
Supermicro AS-5126GS-TNRT2 5U with 10x NVIDIA L40S hardware detail 3

Configuration at a Glance

GPU/Accelerator10x NVIDIA L40S (48GB GDDR6 each)
GPU InterconnectPCIe 5.0 x16, dual-root (up to 13 FHFL slots)
CPU2x AMD EPYC 9005/9004, up to 192C/384T
MemoryUp to 6TB DDR5-6400 ECC RDIMM (24 DIMM slots)

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.

GPU / Accelerator

10x NVIDIA L40S (48GB GDDR6 each)

Processor

2x AMD EPYC 9005/9004, up to 192C/384T

Memory

Up to 6TB DDR5-6400 ECC RDIMM (24 DIMM slots)

Storage

8x 2.5" NVMe + 2x 2.5" SATA hot-swap

Overview

The AS-5126GS-TNRT2 is a 5U dual-socket AMD EPYC system configured with ten double-width NVIDIA L40S GPUs across a dual-root PCIe 5.0 fabric. It is specified, integrated, and validated to a defined inference and rendering profile, then supplied with full manufacturer warranty through authorized distribution.

Who This Solution Is For

Inference platforms serving many concurrent models
Studios running large-scale rendering and VFX
SaaS providers hosting multi-tenant AI endpoints
Teams needing maximum GPU density per node

Business Benefits

Highest PCIe density

Ten double-width L40S cards in one chassis maximize tokens and frames per rack unit.

Dual-root balance

The dual-root PCIe topology splits the GPU complex evenly across both CPUs for consistent latency.

Tested to profile

Each system is benchmarked against your inference or rendering target before sign-off and shipment.

Typical Business Use Cases

1

Concurrent LLM and vision inference

2

Studio rendering and media transcoding

3

Omniverse and digital-twin workloads

4

Multi-tenant AI inference hosting

Industry Applications

Media & EntertainmentSaaS & SoftwareAI & Machine LearningManufacturing

Technical Overview

Based on the Supermicro H14 5U platform with dual AMD EPYC 9005/9004 processors, the chassis presents up to 13 PCIe 5.0 x16 slots in a dual-root layout that hosts ten double-width GPUs plus reserved networking slots. Six 2700W Titanium supplies and ten heavy-duty PWM fans sustain the full L40S complement in air-cooled racks up to 35C.

GPU/Accelerator10x NVIDIA L40S (48GB GDDR6 each)
GPU InterconnectPCIe 5.0 x16, dual-root (up to 13 FHFL slots)
CPU2x AMD EPYC 9005/9004, up to 192C/384T
MemoryUp to 6TB DDR5-6400 ECC RDIMM (24 DIMM slots)
Storage8x 2.5" NVMe + 2x 2.5" SATA hot-swap
Networking2x 10GBASE-T, dedicated 1GbE BMC port
Form Factor5U rackmount, air-cooled (up to 35C)
Power6x 2700W redundant Titanium (3+3 / 4+2)
ManagementASPEED BMC, IPMI 2.0, Redfish, KVM-over-IP

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

Why choose L40S over H100 for this node?

The L40S offers strong inference, fine-tuning, and graphics throughput at lower power per card, so ten of them deliver excellent density for mixed AI and visual workloads.

What does dual-root PCIe mean for performance?

The GPU complex is divided across both CPU sockets, balancing PCIe bandwidth and reducing contention for latency-sensitive inference serving.

Can I reserve slots for high-speed networking?

Yes, the slot layout reserves PCIe lanes for NICs so the node can join a GPU cluster fabric alongside the ten GPUs.

Hardware Assistance

Configure the Supermicro AS-5126GS-TNRT2 5U with 10x NVIDIA L40S 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.