Empowering AI at the Edge will be the focus for Advantech at this year’s embedded world, as the company will showcase its latest breakthroughs in Edge AI computing, including the latest inference systems, high-performance AI server solutions supporting LLM, next-gen AI accelerators, and advanced industrial-grade Qualcomm-based solutions.
Visitors to the company’s booth will be able to see live demos by Advantech experts of the latest entry-, mid- and high-level Edge AI computing solutions. There will also be new products on show developed specifically for autonomous systems and robotics, as well as for electric vehicle (EV) charging applications. Additionally, Advantech will be displaying new enterprise rack-mount server solutions and a range of new accessories such as Wi-Fi 7 modules, Hailo-8 and Intel AI accelerators, plus Compute Express Link (CXL) memory technology.
Advantech’s new Qualcomm-based product portfolio will debut at the event. Key highlights include the Qualcomm embedded industrial-grade mid-tier QCS6490/5430 processors for industrial IoT applications, as well as the QCS9100 with 100 tera-operations per second (TOPS) and Snapdragon X-Elite industrial grade (45 TOPS) processors. The new product range will be available for modules, boards and box-systems and makes the first in a series of innovative developments stemming from Advantech’s strategic alliance with Qualcomm, announced at last year’s show.
Advantech will also showcase its upgraded AIR-150 Edge AI Inference System. Built on the Intel 13th Gen Core-i platform and integrating the high-performance Hailo-8 AI acceleration module (offering 26 TOPS of computing power), the AIR-150 is designed for entry-level to mid-range visual AI recognition tasks. Its low-latency, low-power processing makes it well suited for applications like PPE detection in factory production lines, where it can simultaneously connect two or three cameras to verify that workers are fully equipped.
Also on show will be the latest AIR-120 26 TOPS Edge AI Inference System designed for advanced AI applications. Integrating Intel Alder Lake-N processors (up to four cores and 12W thermal design power) and the rugged AMSTON Lake platform with Hailo-8, the AIR-120 supports AI-driven applications such as autonomous driving, ADAS, industrial automation, and digitised healthcare.
Taking centre stage will be the AIR-520 4U Edge SI server, powered by AMD EPYC-7003 processors with to 64 cores. The server features built-in GenAI Studio, a platform for cost-effective creation, evaluation and integration of custom on-premises large language models (LLMs). Part of Advantech’s Edge AI software development kit (SDK), the AIR-520 is designed to meet growing demand for on-premises LLM applications in manufacturing, healthcare, and retail.
Advantech’s Edge AI SDK with GenAI Studio takes advantage of such AI offerings as OpenAI, Gemini, Anthropic and Ollama. It introduces full-parameter fine-tuning functionality and optimised GPU 4x RTX-4000 Ada generation resources, facilitating high-efficiency AI computing capabilities. Based on quantization technology, the LLM inference model can run seamlessly on Advantech’s AIR-310 MXM GPU Edge AI Inference System.
Finally, Advantech is also advancing real-time computing by integrating Intel Time Coordinated Computing (TCC) solutions. By isolating critical real-time processes from AI workloads, Intel TCC enables high-performance compute processors to be easily configured for supporting high determinism in time to meet user-applications requirements. Intel TCC is one of the key technologies of Intel’s processors for the Edge, supporting use cases across industries such as industrial automation, robotics, retail and avionics.
There’s plenty of other editorial on our sister site, Electronic Specifier! Or you can always join in the conversation by commenting below or visiting our LinkedIn page.
