Geniatech Expands Edge AI Portfolio with New i.MX8MP Series: SoM Modules, SBCs, and Edge AI Systems

Geniatech, a global leader in embedded solutions, today announced a major update to its Edge AI portfolio with the introduction of new products based on the next-generation NXP i.MX 8M Plus processor. The launch includes the SOM-iMX8MP-SMARC module, DB820P SMARC development board, and APC880 edge AI system — further enriching Geniatech’s edge AI portfolio alongside its established Rockchip RK3588 platforms and high-performance AI acceleration modules like AIM-H8 and AIM-M2. Together, these offerings deliver comprehensive, scalable compute solutions for diverse applications, including vision, robotics, and intelligent IoT.​​

Scalable Architecture for Vision, Robotics, and Edge Intelligence

Engineering teams often face challenges when moving from prototype to deployment, including hardware form factor changes, carrier redesign, and software integration barriers. Geniatech addresses these with a reusable platform strategy based on NXP i.MX 8M Plus, enabling a unified CPU architecture and consistent software stack across:

• Edge System (APC880): Industrial-grade, fanless edge AI computer supporting large language models and complex AI integration.
• Single Board Computer (DB820P): SMARC-based compact SBC with 2.3 TOPS NPU and comprehensive I/O, designed for rapid prototyping and reliable deployment.
• System-on-Module (SOM-iMX8MP-SMARC): High-performance, long-lifecycle module ready for customizable embedded designs and multi-board integration.

Geniatech provides full-stack support — reference designs, evaluation kits, and architecture consulting — to help teams make confident technical decisions early in the product cycle and accelerate time-to-market.​

Multi-SoC Collaboration: Enhanced AI Computing Power and System Reliability

NXP i.MX8M Plus Series — Complete Industrial-Grade Edge AI Solutions

• Quad-core Cortex-A53 processor with real-time Cortex-M7, integrated 2.3 TOPS neural network processor, and ISP for multimedia processing.
• Rich industrial I/O (HDMI, MIPI, LVDS, GbE, USB, CAN, RS232/485, Mini PCIe, M.2), with up to 4GB RAM and eMMC options.
• Designed to meet SMARC 2.1 standards and industrial temperature requirements, ensuring seamless migration from concept validation to deployment.​​

RK3588 Series — High-Performance Multimedia AI Solutions

• Octa-core (Quad Cortex-A76 + Quad Cortex-A55) with up to 6 TOPS NPU, 8K video encoding/decoding, up to 32GB RAM and 128GB eMMC.
• Advanced interfaces (HDMI, DP, PCIe, SATA, USB, MIPI) ensure outstanding bandwidth for multimedia and AI workloads in embedded environments.

AI Accelerator Modules — Integrated Hailo-8 and Kinara Ara-2 NPUs

• AIM-H8 (Hailo-8 NPU) delivers up to 26 TOPS; AIM-M2 (Kinara Ara-2 NPU) up to 40 TOPS per module.
• Plug-and-play expansion via M.2/B2B connectors, supporting concurrent multi-stream/multi-model inference for industrial vision and generative AI.
• Optimized for TensorFlow, PyTorch, ONNX compatibility, low latency, and rugged industrial reliability to support current and future edge AI demands.​​

Solution Benefits & Industry Support

Geniatech’s robust, scalable edge AI platforms empower wide-ranging deployments in vision, robotics, smart buildings, and secure IoT gateways. Customers benefit from consistent software, rich multimedia and industrial interface support, and complete engineering resources for fast prototyping and productization. The full portfolio enables secure and reliable deployment of AI and large models at the edge — all designed for industrial durability and future expansion.​

A Geniatech spokesperson commented, “Our new product additions bring greater coverage and deeper technical innovation to our edge AI hardware portfolio, helping clients drive breakthroughs in every phase of the AI development cycle — from R&D to real-world deployment.”

This announcement delivers clear product details, technical depth, and business value, with all claims supported by the latest Geniatech and industry sources. https://www.geniatech.com/