Aetina Corporation has expanded its SuperEdge portfolio for enterprise AI with two compact 2U short-depth GPU servers: the AEX-2UA1, which the company said has achieved NVIDIA Certified System status, and the newly announced AEX-2UA2.
The company said the systems are aimed at enterprise edge and near-edge deployments where space constraints, data sovereignty requirements, latency and installation flexibility can limit the use of conventional full-depth data centre servers or cloud-only approaches. Aetina positioned the servers for on-premises generative AI, AI agents and inference workloads.
“Enterprise AI is moving from experimentation toward production, where AI agents and generative models need reliable infrastructure that operates close to enterprise data,” said Jame Su, Software Division Director of Aetina. “With the AEX-2UA1 achieving NVIDIA Certified System status and the addition of the AEX-2UA2, we’re giving enterprises more flexibility to match GPU performance and deployment format to their AI workloads, helping them move from development to production with confidence.”
Aetina said it is a system vendor acknowledged by NVIDIA to design MGX servers based on NVIDIA’s MGX modular reference architecture. The company framed the AEX-2UA1 and AEX-2UA2 as standardised infrastructure building blocks intended to align with the NVIDIA AI software stack, lower integration risk and speed production rollouts.
According to Aetina, the AEX-2UA1 achieved NVIDIA Certified System status when configured with two NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, and is also qualified for NVIDIA L40S and NVIDIA H200 NVL GPUs. The company said the server is powered by a single-socket Intel Xeon 6700-series processor with P-cores and supports dual double-width GPUs up to 600W, with NVIDIA NVLink for H200 NVL configurations.
Aetina said the AEX-2UA1 also includes a dedicated PCIe Gen5 x16 slot for an NVIDIA BlueField-3 DPU or ConnectX-7 NIC, up to four hot-swappable PCIe Gen5 E1.S bays, and two PCIe Gen5 M.2 slots with NVMe RAID 0/1/5/10 support. The company listed the chassis dimensions as 438 × 420 × 88 mm and pointed to deployment scenarios including enterprise IT rooms, smart manufacturing sites, telecom environments, retail facilities and AI innovation centres.
The AEX-2UA2, Aetina said, adds an alternative rear-access 2U short-depth architecture to the SuperEdge line. The company said it pairs an Intel Xeon 6700P processor with a broader range of GPU configuration options, targeting enterprise and systems integrator requirements such as rack layouts, cabling and maintenance workflows.
Aetina also described use cases including private AI knowledge assistants and executive decision intelligence deployed entirely on-premises. It said configurations could range from semantic search and departmental Q&A on NVIDIA L40S to running larger models on NVIDIA H200 NVL and RTX PRO 6000 Blackwell GPUs, including fine-tuning with proprietary data using dual H200 NVL GPUs connected via NVLink.
