Prev
Next
AEX-2UA1
• Single Intel® Xeon® 6700-series with P-cores.
• 2U server - 420mm (16.5") short-depth rackmount server.
• 2 x PCIe Gen5 x16 slots (FHFL) and 1 x PCIe Gen5 x8 slot (FHFL) – supporting up to 2 double-width PCIe GPUs.
• 1 x PCIe Gen5 x16 slot (FHHL) - supporting NVIDIA® BlueField® -3 or NVIDIA ConnectX® -7.
• Up to 4x Hot-Swap PCIe Gen5 E1.S and 2x PCIe Gen5 M.2 slots.
• 1+1 2000W Redundant Titanium level power supply.
Overview
Accelerate On-Prem LLM Deployment with SuperEdge AEX-2UA1
Short-Depth GPU Edge AI Server Built on NVIDIA MGX™ Architecture
Transform enterprise operational efficiency with SuperEdge AEX-2UA1—a compact, high-performance GPU server delivering accelerated generative AI for enterprise and industrial applications. Purpose-built for the edge where data privacy, ultra-low latency, and space constraints are mission-critical, AEX-2UA1 brings edge data center-class AI computing closer to where data is generated.
SuperEdge™ AEX-2UA1 integrates a single-socket Intel® Xeon® 6700-series processor with Performance-cores and supports dual double-width, high-performance GPUs via NVIDIA NVLink™, providing exceptional AI performance and 24/7 reliability for on-premises large language model (LLM) and multimodal inference and training, as well as design simulation and rendering workloads. By bridging edge deployments and centralized data centers, the server optimizes data flow and operational efficiency while delivering a cost-effective, enterprise-grade supercomputing platform for workflow AI agents, LLM assistants, and digital humans across industries such as smart manufacturing, retail, and enterprise IT environments.
Built on the NVIDIA MGX™ modular architecture and certified as an NVIDIA Certified System (NCS), SuperEdge AEX-2UA1 leverages a scalable, flexible, architecture, future-proofed for next generation NVIDIA Blackwell GPUs with NVLink™ and BlueField® -3 DPU. Encased in a 2U short-depth form factor, AEX-2UA1 is optimized for enterprise IT environments, and industrial deployments across standard racks, short-depth racks, or non-rack edge locations. AEX-2UA1 provides maximum compute density and flexibility without the complexity or space requirements of traditional data center infrastructure.
Single Intel Xeon 6700-series with P-cores
NVIDIA MGX™ Modular Server Design
Dual Double-Width NVIDIA GPUs with NVLink
High-Speed Networking
2U Short-Depth Footprint
Enterprise AI Training and Inference
| ● NVIDIA-Certified System for Confidence: NCS with NVIDIA L40S, NVIDIA RTX PRO 6000 Blackwell Server Edition, H200NVL GPUs |
| Featuring NVIDIA L40S, NVIDIA RTX PRO 6000 Blackwell Server Edition, H200NVL GPUs, SuperEdge AEX-2UA1 meets rigorous standards of performance and reliability of NCS certification. |
| ● Unmatched On-Prem Gen AI Performance：Data-Proven Incredible Visualization and LLM inference |
| SuperEdge AEX-2UA1 leverages one single-socket Intel® Xeon® 6700-series processor with P-cores and dual double-width, high-performance NVIDIA GPUs with NVIDIA NVLink™. |
| ● NVIDIA MGX™ Modular Architecture for Scalability and Flexibility |
| Built on NVIDIA MGX™, AEX-2UA1 offers modular flexibility and future-ready compatibility, enabling faster deployment, optimized AI acceleration, and seamless scaling as AI workloads evolve. |
| ● Reliable, Space-Efficient Server Design for 24/7 Operation |
| Hot-swappable fans, E1.S storage, and power supplies enable tool-free maintenance and zero downtime. Combined with 1+1 redundant power supplies and intelligent BMC management, the system delivers stable, continuous performance for mission-critical environments. Its space-saving, compact 2U short-depth form factor supports both rack-mounted and standalone deployments, providing maximum flexibility across edge and enterprise installations. |
| ● Accelerated Edge AI for Industrial AI and Enterprise AI |
| Optimized for on-premises generative AI, AEX-2UA1 enables secure local processing of sensitive data for privacy-critical sectors such as finance, healthcare, smart manufacturing, and space-constrained 5G telco environments, offering a highly efficient entry-level LLM deployment alternative to traditional data center GPU servers. |
Supercomputing Edge Server for Enterprise and Industrial AI
Unlock next-level GenAI performance with SuperEdge™ AEX-2UA1, featuring double-deck high-performance GPUs — including NVIDIA L40S, NVIDIA RTX PRO 6000 Blackwell Server Edition, and H200 NVL. Designed for real time LLM and VLM computing at the edge.
Compatible PCIe series
-
NVIDIA L40S
-
NVIDIA RTX PRO 6000 Blackwell Server Edition
-
NVIDIA H200 NVL
Specifications
| Model Number | AEX-2UA1 | |
|---|---|---|
| Chassis | Form Factor | Front Access 2U Rackmount |
| Dimensions (W x D x H) | 438mm x 420mm x 88mm | |
| Processor | Single Intel® Xeon® 6700-series with P-cores | |
| Chipset | N/A | |
| GPU | Max Count | Up to 2 Double-width PCIe Passive GPUs |
| Qualified | L40S, RTX PRO™ 6000 Blackwell Server Edition, H200NVL (with NVLink bridge) | |
| System Memory | 8 x Channel Memory | |
| Storage | Drive Bays | Rear: 4x Hot-swappable E1.S, up to PCIe Gen5, up to 15mm E1.S |
| M.2 | 2 x M.2, up to PCIe Gen5 x4, M-Key 2280/22110 | |
| RAID Support | NVMe RAID 0/1/5/10 support (Intel® VROC RAID key required) | |
| Input/Output | 1 x RJ45 1Gbe Dedicated BMC LAN port 1 x MiniDP port 2 x USB 2.0 ports |
|
| Front Panel | Buttons | System Power UID/BMC reset |
| LED | Power status UID Storage Activity Network Activity Information |
|
| PCIe Expansion slots | 2 x PCIe Gen5 x16 link (PCIe x16 slot, FHFL) for 2 Double-width GPUs with NVLink bridge 1 x PCIe Gen5 x8 link (PCIe x16 slot, FHFL)* 1 x PCIe Gen5 x16 link (PCIe x16 slot, FHHL) *Slot not avaliable when 2 Double-width AOC is installed. |
|
| Power | Power Supply | Default: 2 x 2,000W Redundant (1+1) Titanium Level (96%) power supplies |
| Power Input | Default 2,000W power supply: 1,000W: 100-127V~, 12A, 47-63Hz 2,000W: 220-240V~, 10A, 47-63Hz |
|
| Power Connectors | IEC 60320 C13 | |
| Package | TBD | |
| Weight | TBD | |
| MISC. Function | Trusted Platform Module (TPM) 2.0 Intel® Platform Firmware Resilience (Intel® PFR) |
|
| Temperature | Operating Temperature: 10°C ~ 35°C (50°F ~ 95°F) Non-operating Temperature: -30°C to 60°C (-22°F to 140°F) |
|
| Humidity | Operating Relative Humidity: 8% to 80% (non-condensing) Non-operating Relative Humidity: 8% to 90% (non-condensing) |
|
| OS Support | TBD | |
| Certification | TBD | |
Specifications
Dimensions
Ordering Information
| Model Number | Description |
|---|---|
|
AEX-2UA1 |
AEX-2UA1 2U GPU server (Bare system, w/ Xeon 6731P CPU, w/o RAM, SSD, GPU Card) |
Downloads
| Model Name | File Name | OS | Version | Update Date | Download |
|---|---|---|---|---|---|
| AEX-2UA1 | Aetina_AEX-2UA1_Datasheet | v1.0 | 2026-06-08 | Download |