Aetina Partners With Hailo to Launch Multi-Inference AI Solutions at the Edge During Isc West 2023

Aetina and Hailo join forces to launch multi-inference AI solutions that involve object recognition AI models, Aetina’s AI inference platform—AIP-SQ67, and AI-MXM-H84A MXM that is built with 4x Hailo-8™ AI accelerators. Empowered by Aetina’s AI-MXM-H84A MXM module that provides up to 104 Tera-Operations Per Second (TOPS) of AI performance from Hailo’s AI processors, the AIP-SQ67 platform delivers enough computing power to enable real-time video analytics processing and multiple, low-latency AI inference tasks at the edge.

Recommended AI: The Future of AI Is Here. Now Let’s Make It Ethical

The AI solutions are capable of detecting different objects including humans, vehicles, and analyzing many video datasets from several cameras at the same time, which makes the solutions ideal for various applications in cities and transportation systems. The solutions will be showcased by Aetina and Hailo during ISC West 2023.

Aetina’s MegaEdge AI Inference Platforms

AIP-SQ67—part of Aetina’s MegaEdge family—features an Intel 12th Gen Core processor, and has expansion slots for one MXM and up to two M.2 AI accelerators. The x86 AI computer, designed for high velocity data analytics, has different types of I/O configuration accommodating a wide range of peripherals and sensor devices, enabling the platform to cater a broad spectrum of AI verticals.

Recommended AI: AiThority Interview with Jessica Kipper, Senior Director, Software Product Management at Schneider Electric

MXM Empowered by Hailo-8™ edge AI Processors

The AI-MXM-H84A MXM is designed with four Hailo-8 edge AI processors; each of the processors provides up to 26 TOPS of AI performance. With an architecture that takes advantage of the core properties of neural networks, the edge AI processor allows edge devices to run deep learning applications at full scale more efficiently, effectively, and sustainably. The AI-MXM-H84A MXM, due to its small-sized form factor, can be easily integrated into a variety of embedded systems to handle heavy inference workloads.

Recommended AI: AiThority Interview with Pete Wurman, Director at Sony AI

[To share your insights with us, please write to sghosh@martechseries.com]