AI-powered edge computing innovations unveiling

Aetina is unveiling a robust portfolio of edge AI hardware and software solutions at embedded world 2025, empowering businesses worldwide to unlock new opportunities at the intelligent edge.

Leading the lineup is the company's DeviceEdge series, powered by NVIDIA Jetson Orin SoMs, including the new NVIDIA Jetson Orin Nano Super and Jetson Orin NX Super. These high-performance devices are designed to support conventional edge AI applications and cutting-edge generative AI workloads, allowing dynamic and context-aware decision-making processes at the edge across industries such as agriculture, transportation, and smart motion analysis. As an Elite Partner of NVIDIA, the company offers a versatile range of compact devices powered by NVIDIA Jetson modules, including the Jetson AGX Orin, Orin NX, and Orin Nano. It works closely with software vendors and system integrators to deliver tailored AI applications.

The company is collaborating with Qualcomm Technologies, Inc. to deliver powerful AI capabilities, including generative AI and computer vision, for local, on-premises deployments using the Qualcomm AI On-Prem Appliance Solution and Qualcomm AI Inference Suite. The flagship MegaEdge AIP-FR68 integrates two Qualcomm Cloud AI 100 Ultra cards, with each card offering an impressive 870 TOPS of AI computing at just 150W. Supporting dual AI cards in a single workstation, this system redefines enterprise-grade AI processing with unmatched efficiency and scalability, providing unprecedented computing power and ready-to-use AI applications for enterprises and industrial organisations. Customers can now employ generative AI using their proprietary data, fine-tuned models, and technology infrastructure to automate human and machine processes and applications in virtually any end environment.

The company is also presenting its collaboration with Axelera for computer vision solutions, featuring the MegaEdge AIP-MURE series powered by 13th Gen Intel Core processors and Axelera AI M.2 AI Edge accelerator cards. This versatile, compact system delivers up to 300 TOPS, optimised for smart surveillance and real-time video analysis applications.

An ARM-based AIB-MR1A-A1 edge device, equipped with the Rockchip RK3588 8-Core processor and support for 2280 M.2 AI accelerators like Axelera, EdgeCortix, and Hailo is being introduced by the company. This platform allows powerful AI processing, making it excellent for real-time video analytics and edge AI inference. Its PCIe Gen3 x4 M.2 slot ensures high-speed data transfer, making it appropriate for applications demanding swift and efficient processing.

Also being presented is the SuperEdge AEX-2UA1, a compact and advanced short-depth edge server based on NVIDIA MGX modular reference design, coupled with Intel Xeon processors. This server is engineered for environments where space is limited but computational demands are high. The server will also feature a live demo of a Private LLM deployed directly on the AEX-2UA1. This LLM uses Retrieval-Augmented Generation (RAG) technology to provide secure and highly specialised AI capabilities tailored to specific domains. By running locally on the device, this setup ensures data privacy and reduced latency, making it an excellent solution for scenarios where secure, domain-focused AI processing is essential.

These innovative Edge AI solutions can be viewed at embedded world 2025, Booth 1-370, where you can experience live demonstrations and engage with experts about the latest advancements in AI hardware and applications.