IEI Offers Future-Ready Edge AI Platform in the TANK-XM813, Powered by Intel

By Ken Briodagh

Editor in Chief

Embedded Computing Design

October 22, 2025

Blog

Image Credit: IEI Integration

The drive to move processing to the Edge, where data is collected, has been ongoing for decades, but the rise of AI tools has made intelligence at the Edge a mission-critical development path for any industrial enterprise.

To accomplish the kind of AI-ready compute at the Edge demanded for industrial applications, devices need to combine powerful processing, resiliency, and robustness in challenging environments, energy efficiency, flexible memory, and adaptable features. All these elements working together and engineered into a single solution is how Edge AI will be possible now and in the future.

IEI Integration has created such a device in its TANK-XM813. This Edge AI platform has been designed to be ready for the Industrial IoT demands of today and the future, according to IEI.

The TANK-XM813 is engineered to deliver high-performance, reliable computing at the rugged edge. The embedded compute platform is powered by an Intel® Core™ Ultra Processor and supports expandable GPU capability, high-speed NVMe storage, and scalability through IEI’s eChassis modules.

The platform achieves Edge AI acceleration through the combination of Intel® Core™ Ultra processors, NPU, and GPU, which produce the flexible and reliable computing necessary to meet industrial demand for real-time AI at the Edge. If more compute is desired or required, the platform also supports a modular chassis with up to six expansion slots for GPU or I/O flexibility.

To meet the challenges of robustness and resiliency, the IEI TANK-XM813 has a fanless design with wide temperature tolerances of -20°C to 60°C (-4°F to 140°F) and 12–28V DC input. These hardware design features mean the TANK-XM813 can ensure stable performance in harsh environments.

It supports up to 96GB of DDR5 memory, has dual 2.5GbE LAN, and 8 optional 2.5GbE LAN ports with Power over Ethernet (POE) via an expansion module. These memory and network options enable the TANK-XM813 system to provide seamless data transmission and AI device integration.

IEI says the TANK-XM813, when compared with conventional embedded PCs, can deliver higher AI efficiency, lower latency, and lower power consumption through heterogeneous computing than almost any competitive device.

The applications for an Edge platform like this one are vast. It’s ideally suited to machine vision, industrial automation, and intelligent control at the edge, but that’s just the beginning. Smart surveillance, intelligent transportation, and advanced manufacturing are all on the roadmap for the TANK-XM813 and, in some cases, are already in the field.

Its modularized design and easy-to-use configuration tools mean that the system can be installed and deployed quickly and efficiently, reducing lead times, potential downtime, and inventory costs.

In the industrial embedded computing space, users are as varied as the applications that can be imagined. That’s why IEI built the TANK-XM813 to be modular and flexible to meet a wide range of edge AI computing needs, ensuring faster deployment and greater operational efficiency, even when deploying for a custom application.

Intel’s AI Edge Initiative

This blog is part of a series showcasing Intel’s AI Edge initiative, designed to highlight the latest innovations of AI and edge computing. Intel recently unveiled its Intel® AI Edge Systems, Edge AI Suites and Open Edge Platform. These solutions are designed to integrate AI into partners’ existing infrastructure, kickstarting development to enhance system reliability and strengthening security.

Intel is co-innovating with its software partners in AI creation and optimization for edge applications, as illustrated by this series of blog posts. Furthermore, Intel is driving innovation alongside its hardware platforms to optimize AI Edge systems for key workloads, offering best fit performance across a range of power levels, sizes and performance options.

To find out more, click here to visit Intel.

Ken Briodagh is a writer and editor with two decades of experience under his belt. He is in love with technology and if he had his druthers, he would beta test everything from shoe phones to flying cars. In previous lives, he’s been a short order cook, telemarketer, medical supply technician, mover of the bodies at a funeral home, pirate, poet, partial alliterist, parent, partner and pretender to various thrones. Most of his exploits are either exaggerated or blatantly false.

More from Ken