Supermicro has expanded its edge AI systems portfolio with Intel-powered platforms designed for low-latency inference, industrial automation, and local processing in distributed environments.The expanded range includes systems using Intel Core Ultra Series 3 processors, Intel Core Series 2 processors, and Intel Arc Pro B-series GPUs. The systems cover compact fanless units, short-depth 1U rackmount servers, and a mini tower for office and edge deployments.Supermicro said the systems are aimed at organisations in retail, manufacturing, physical security, transportation, and logistics. The systems are intended for edge environments involving cameras, sensors, machines, scanners, and other connected equipment.Edge systems expandIDC estimated global spending on edge computing solutions at nearly $261 billion in 2025, with spending forecast to reach nearly $380 billion by 2028.The systems also sit within a wider connected-device environment. IoT Analytics estimated that the number of connected IoT devices would reach 39 billion by 2030, reflecting a 13.2% compound annual growth rate from 2025.Edge computing processes data closer to devices and local data sources. Edge systems can also support local processing in environments where cloud connectivity is limited or intermittent.Industrial IoT deployments can involve continuous streams of operational data from connected equipment. In manufacturing, these workloads can include computer vision, machine monitoring, production-line inspection, and automation tasks.Physical security deployments can involve video analytics near camera networks. Logistics and transportation sites can involve scanners, cameras, tracking infrastructure, and other local data sources.Fanless and compact systemsThe fanless SYS-E103-14P is one of the industrial edge systems in the expanded portfolio. It uses Intel Core Ultra Series 3 processors and is built for AI inference workloads such as computer vision and industrial automation.The system is compact and DIN-rail mountable, which allows it to be installed in industrial control cabinets or space-constrained edge locations. Supermicro said it includes an integrated GPU and neural processing unit, delivering up to 180 TOPS of combined AI performance without requiring a discrete GPU.These form factors support deployments in control cabinets, equipment rooms, branch sites, and constrained industrial spaces.TOPS, or trillion operations per second, is commonly used as a hardware comparison metric for AI processors and accelerators. Actual performance can vary depending on workload, model type, configuration, software, and system design.The SYS-E103-14P supports up to 128GB of DDR5 memory and includes multiple I/O options. Supermicro said the system supports operating temperatures from 0°C to 45°C.Workloads such as video analytics, machine vision, access control, and equipment monitoring rely on data from cameras, sensors, storage, and networks, making memory, bandwidth, and I/O relevant system requirements.Supermicro also introduced the SYS-521AD-LN2, a slim AI mini tower powered by Intel Core Series 2 processors. The system supports up to 12 high-performance P-cores, up to 64GB of DDR5 memory, and compact GPU accelerators.The mini tower is designed for local AI inference, model development, and fine-tuning in office and edge environments. Supermicro said it supports accelerators including the Intel Arc Pro B50 GPU and NVIDIA RTX Pro Blackwell 2000 GPU.Supermicro positions the broader portfolio around edge inference, while the SYS-521AD-LN2 mini tower is also designed for model development and fine-tuning.Supermicro has also updated two existing edge systems. The short-depth 1U SYS-111AD-WN2R and compact SYS-E300-13AD5 now support Intel Core Series 2 processors.The updates allow users to increase AI and compute performance while keeping existing deployment footprints. Both systems also support DDR5 memory.Short-depth rackmount systems can be used in branch sites, retail locations, telecom rooms, industrial facilities, and other constrained spaces.Intel GPUs add local AI accelerationThe expanded portfolio also adds broader support for Intel Arc Pro B-series GPUs across Supermicro’s edge AI server range. These GPUs provide discrete acceleration for AI and visual computing workloads.Intel says its Arc Pro B-series GPUs combine dedicated graphics memory, XMX AI engines, and multi-GPU support for rendering, video processing, and AI workloads. These capabilities apply to workloads such as image analysis, video processing, and local inference.The Intel Arc Pro B70 offers up to 367 TOPS and up to 32GB of VRAM. The Intel Arc Pro B60 delivers up to 197 TOPS and supports expanded memory bandwidth and multi-GPU configurations. The lower-power Intel Arc Pro B50 provides up to 170 TOPS for smaller edge systems and workstations.Supermicro’s updated systems use different combinations of CPUs, integrated graphics, NPUs, and discrete GPUs across compact and larger edge platforms.Deployment and managementSupermicro said its Data Center Building Block Solutions portfolio supports modular deployments using validated components and subsystems. The portfolio covers individual servers, networking, rack-scale systems, data centre-level systems, software, and services.Mory Lin, vice president of IoT, embedded, and edge computing at Supermicro, said organisations need edge infrastructure that can support real-time inference, low-latency performance, and power efficiency near where data is generated.Dan Rodriguez, corporate vice president and general manager of the Edge Computing Group at Intel, said edge AI workloads require compute performance, power efficiency, scalable acceleration, and cost considerations. He said the combination of Intel Core Ultra processors, Arc Pro GPUs, and Supermicro’s edge systems is intended to support AI deployments across real-world environments.The announcement adds Intel-based options across several edge form factors, from fanless industrial systems to compact towers and short-depth rackmount servers.(Photo by BoliviaInteligente)See also: NVIDIA Halos OS upgrades the safety of physical AI workloadsWant to learn more about IoT from industry leaders? Check out IoT Tech Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information.IoT News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.