With new standards for AI inference, Axelera Metis AI accelerators now at Rutronik

Rutronik has recently added Axelera AI, a European pioneer in AI and edge computing, to its franchise partners. Axelera Metis products offer an alternative to traditional GPU-based solutions. The Metis AI platform is a holistic AI solution that combines state-of-the-art hardware and software. The hardware portfolio includes the Metis M.2 AI Inference Acceleration Card and the Metis PCIe AI Inference Acceleration Card.

At the heart of the Metis AI platform is the Metis AIPU (AI Processing Unit), a quad-core chip based on a proprietary RISC-V architecture and Digital In-Memory Computing (D-IMC). This architecture enables AI inference performance of up to 214 TOPS with energy efficiency of 15 TOPS/W. This makes it ideal for computer vision inference on the edge and provides scalable technology to meet the future demands of AI acceleration workloads.

The hardware is complemented by the Voyager SDK, a software package that simplifies model deployment, optimisation and integration.

Metis M.2 AI Inference Acceleration Card
This card, powered by the Metis AIPU, is ideal for applications with limited space that require powerful AI inference. In addition, the M.2 card has 1 GB of dedicated DRAM memory, which helps reduce power consumption and enables easier integration into systems. Its inference performance is up to 214 TOPS with 5 to 9 W typical application performance. A PCIe Gen3 x4 interface and security features (Secure Boot, Root of Trust) round out the product. The Metis M.2 AI Inference Acceleration Card is optionally available with an active cooling solution.

Target applications:
• Surveillance: easy deployment in buildings, construction sites or event venues
• Robotics: high-resolution advanced gesture recognition, pose estimation and decoding of human
intentions
• Medical: improved accuracy of real-time diagnosis, monitoring, imaging and analysis
• Agritech: crop analysis, smart irrigation, pest control, automated harvesting
• Drones: navigation, object detection and monitoring in lightweight, battery-powered UAV systems

Metis PCIe AI Inference Acceleration Card
The PCIe single slot cards in HHHL form factor with PCIe Gen3 x4 interface are also powered by the Metis AIPU. The PCIe version features up to 3,200 FPS for ResNet- 50, up to 214 TOPS with typical application performance of 8 W to 15 W, and active air cooling. Security features such as Secure Boot and Root of Trust are also integrated. It is available with 4 GB and 16 GB memory configurations.

Additional options with four Metis chips will be available soon, enabling up to 856 TOPS. These will then be available with 16 GB and 64 GB memory.
Target applications:
• Security: faster detection of incidents such as break-ins, falls or abandoned luggage
• Industry 4.0: fault detection, quality control, occupational safety
• Retail: customer behaviour, inventory monitoring, automated checkout systems
• Mobility: toll collection, enforcement and parking management
• Logistics: improving operational efficiency and safety in logistics centres

Both products support simultaneous inference on multiple cameras and enable the parallel use of independent neural networks. They offer a wide range of end-to-end AI pipelines and pre-trained models for a wide variety of applications. Advanced quantisation tools also ensure high prediction accuracy without compromising performance.

https://rutronik-tec.com

> Read More

Ibase announce new ultra-compact rugged embedded edge computer

IBASE Technology has released the CP100 ultra-compact rugged embedded edge computer, purpose-built for high-reliability and long-term operation in space-limited environments. Powered by the AMD Ryzen Embedded R2000 Series processor, the CP100 offers three chipset configurations, allowing users to tailor computing performance to specific application needs.

Engineered for minimal footprint, the CP100 adopts a 2.5-inch SBC design with a compact system size of just 94(W) x 133(D) x 51(H)mm, making it ideal for industrial and embedded edge applications requiring efficient, multi-threaded performance in tight spaces. It also supports DIN-rail mounting, enabling easy deployment in various automation systems and equipment architectures.

The CP100 features dual HDMI 2.0 ports, supporting simultaneous dual 4K display output for consistent, high-quality video performance. It supports PDPC (Peripheral Device Power Control) function, optimising power efficiency for connected peripherals. With four configurable RS-232/422/485 serial ports, the system enables seamless connectivity with a wide range of industrial sensors and devices, making it well-suited for smart manufacturing, industrial automation, smart retail, and intelligent logistics applications.

To meet the demands of high-speed networking and flexible storage expansion, the CP100 includes multiple M.2 slots (E-Key and M-Key), supporting WiFi, Bluetooth, LTE/5G wireless modules, and high-performance NVMe storage devices, delivering scalability and future-proofing for evolving system requirements.

Key Features of CP100:

• Powered by AMD Ryzen Embedded R2000 processors with scalable performance options
• Ultra-compact design with DIN-rail support for easy deployment in tight spaces
• Dual HDMI 2.0 outputs for simultaneous 4K video display
• Four configurable COM ports for versatile industrial device integration
• Multiple M.2 expansion slots for wireless connectivity and high-speed storage

http://www.ibase.com.tw

> Read More

Renesas introduces 64-bit RZ/G3E MPU for high-performance HMI systems

Renesas has announced the launch of its new 64-bit RZ/G3E microprocessor (MPU), a general-purpose device optimised for high-performance Human Machine Interface (HMI) applications. Combining a quad-core Arm® Cortex-A55 running at up to 1.8GHz with a Neural Processing Unit (NPU), the RZ/G3E brings high-performance edge computing with AI inference for faster, more efficient local processing. With Full HD graphics support and high-speed connectivity, the MPU targets HMI systems for industrial and consumer segments including factory equipment, medical monitors, retail terminals and building automation.

At the heart of the RZ/G3E is a quad-core Arm Cortex-A55, a Cortex-M33 core, and the Ethos-U55 NPU for AI tasks. This architecture efficiently runs AI applications such as image classification, object recognition, voice recognition and anomaly detection while minimising CPU load. Designed for HMI applications, it delivers smooth Full HD (1920×1080) video at 60fps on two independent displays, with output interfaces including LVDS (dual-link), MIPI-DSI, and parallel RGB. A MIPI-CSI camera interface is also available for video input and sensing applications.

The RZ/G3E is equipped with a range of high-speed communication interfaces essential for edge devices. These include PCI Express 3.0 (2 lanes) for up to 8Gbps, USB 3.2 Gen2 for fast 10Gbps data transfer, and dual-channel Gigabit Ethernet for seamless connectivity with cloud services, storage, and 5G modules.

Starting with the third-generation RZ/G3S, the RZ/G series includes advanced power management features to significantly reduce standby power. The RZ/G3E maintains sub-CPU operation and peripheral functions while achieving low power consumption around 50mW and around 1mW in deep standby mode. It supports DDR self-refresh mode to retain memory data, enabling quick wake-up from deep standby for running Linux applications.

Renesas continues to offer the Verified Linux Package (VLP) based on the reliable Civil Infrastructure Platform, with over 10 years of maintenance support. For users requiring the latest versions, Renesas provides Linux BSP Plus, including support for the latest LTS Linux kernel and Yocto. Ubuntu by Canonical and Debian open-source OS are also available for server or desktop Linux environments.

Renesas has also introduced system-on-module (SoM) solutions featuring the RZ/G3E. A broad range of SoM solutions will be available from Renesas’ ecosystem partners such as a SMARC module from Tria, an OSM (Size-M) from ARIES Embedded, and an OSM (Size-L) from MXT.

https://www.renesas.com/rzg3e

> Read More

Adlink launches cExpress-R8 a compact solution to Demanding Industrial Workloads

Adlink has introduced the cExpress-R8, a COM Express COM.0 R3.1 Type 6 Compact module powered by AMD Ryzen Embedded 8000 Series processors (8845HS, 8840U, 8645HS, 8640U). Engineered for the demands of modern smart industrial applications, it delivers an advanced integration of high-performance and energy-efficient design. This module ignites mission-critical AI capabilities, accelerated edge computing, and advanced graphics performance.

Pushing the limits of edge performance, Adlink’s cExpress-R8 combines up to 8 “Zen 4” cores and 16 threads with integrated AMD RDNA 3 graphics and an XDNA NPU, delivering up to 40 TOPS. The cExpress-R8 fuels power-efficient AI inferencing and fast decision-making across machine vision, robotics, medical imaging, infotainment, and intelligent human-machine systems. With XDNA architecture, performance-per-watt efficiency is optimised, delivering advanced AI processing, and enabling more complex AI workloads without a significant spike in energy demands.

To guarantee real-time processing, the cExpress-R8 supports up to 96GB of DDR5 memory at 5600 MT/s, with both ECC and non-ECC options available. ECC acts as a watchdog, detecting and correcting errors in environments, where failure isn’t an option, whether in mission-critical applications, gaming, transportation, and POS (Point of Sale)/POI (Point of Information) systems.

With support for up to four displays, DP, eDP, HDMI, LVDS, and VGA interfaces. VGA and eDP outputs are available as optional features, allowing flexible visual configurations to meet edge computing demands. Combined with scalable performance, power, and graphics configurations, the cExpress-R8 is well-suited for a diverse range of edge applications. With 10-year product availability, it ensures long-term design stability and lifecycle alignment for system integrator roadmaps.

Adlink’s cExpress development kits, including reference carriers and full I/O support, will be available in Q4 of 2025 to help you accelerate prototyping and simplify system integration.

https://www.adlinktech.com

> Read More

About Smart Cities

This news story is brought to you by smartcitieselectronics.com, the specialist site dedicated to delivering information about what’s new in the Smart City Electronics industry, with daily news updates, new products and industry news. To stay up-to-date, register to receive our weekly newsletters and keep yourself informed on the latest technology news and new products from around the globe. Simply click this link to register here: Smart Cities Registration