Industrial PC maker IEI Integration joined Castec International at AMD Advancing AI 2026 in San Francisco on July 21 and 22 to showcase the new NANO-X100 4-inch single-board computer and edge AI inference applications, underscoring the two companies' collaboration on integrated edge-computing hardware and software.
IEI said the demo focused on autonomous mobile robot (AMR) controllers and smart video surveillance applications. The NANO-X100, built on AMD Ryzen AI Embedded X100 series processors, works with IEI's in-house AiEi software platform to deliver a complete architecture spanning hardware compute, image analysis, and event management.
As edge AI moves beyond simple data processing toward scenarios that require real-time response and decision-making, IEI said its showcase emphasized software-hardware integration rather than raw hardware performance alone.
IEI vice general manager Yao-Tsung Lee said customers deploying edge AI now need more than computing power. They also need systems that can analyze images on site in real time, define events, trigger alerts, and respond quickly.
By presenting the NANO-X100 together with the AiEi software platform, IEI said it hopes to help system integrators accelerate the transition to the next stage of edge AI smart surveillance applications.
On the hardware side, the NANO-X100 uses AMD Ryzen AI Embedded X100 series processors that integrate CPU, GPU, and NPU cores. Its dedicated AI inference performance reaches up to 50 TOPS, and it supports up to 128GB of memory to handle demanding edge workloads such as image processing and vision language models (VLMs).
The platform also offers dual 2.5GbE Ethernet ports, USB 3.2, and M.2 NVMe expansion interfaces, allowing system integrators to combine cameras and peripheral devices based on different application scenarios.
On the software side, IEI said its AiEi platform was another highlight of the showcase.
Through a centralized web dashboard, the platform lets users set monitoring conditions with intuitive text commands, such as checking whether workers are wearing safety helmets and reflective vests, or detecting whether people enter restricted machine zones. When anomalies occur, it can automatically capture images, analyze video, and send alerts by email or text message, turning traditional passive monitoring into active event management.
AiEi currently supports multiple AI models, including object detection, vision-language models, and image enhancement. Looking ahead, IEI plans to connect third-party VLM platforms through an access token mechanism, as well as support integration with vision-language-action (VLA) models and ROS/ROS 2 robotic systems, gradually extending its applications from video surveillance to smart manufacturing, industrial safety, and facility security.
IEI also said it will continue working with AMD and ecosystem partners to push edge AI deployment across more industries.
Article translated by Lily Hess and edited by Jack Wu
