Artificial intelligence chip company Hailo Technologies Ltd. said today it’s launching two new acceleration modules that will boost the processing capabilities of edge devices that run its specialist hardware.

Hailo burst onto the AI scene in 2019 with a customized processor for running deep learning workloads at the edge of the network. The company, which is primarily focused on the automotive sector, said at the time that its Hailo-8 Deep Learning chip enables devices such as autonomous vehicles, smart cameras, drones and AR/VR platforms to run sophisticated deep learning applications at the edge that could previously be hosted only in cloud data centers.

The Hailo-8 processor, which is smaller than a penny, was built from the ground up with completely redesigned memory, control and compute architecture components that enable “higher performance, lower power and minimal latency.” Hailo also provides a software development kit for developers to build apps customized for the hardware.

Hailo is well-funded, having raised $60 million in its most recent Series B funding round in May.

The company said its new M.2 and Mini PCIe high-performance AI acceleration modules support popular deep learning frameworks such as Tensorflow and the Open Neural Network Exchange. They can be integrated with its Hailo-8 processor to accelerate numerous types of deep learning applications, including image classification, object detection, segmentation, pose estimation and more.

Hailo makes some big claims about the new modules’ capabilities. It provides the example of “fanless AI edge boxes,” which are used to connect numerous cameras or sensors to a single intelligent processing device at the edge. It said that the modules enable the Hailo-8 processor to deliver up to 26 tera-operations per second, with a power efficiency of 3 TOPS/W.

Hailo said the module-accelerated Hailo-8 chip compares well with rival processors on several common neural network benchmark tests. For example, it achieved a 26-times higher frames per second rate than Intel’s Myriad-X1 modules, and a 13-times higher rate than Google’s Edge Tensor Processing Unit modules.

Hailo Chief Executive Orr Danon said the new modules will prove to be a game changer for devices at the edge.

“Manufacturers across industries understand how crucial it is to integrate AI capabilities into their edge devices,” Danon said. “The new Hailo-8 M.2 and Mini PCIe modules will empower companies worldwide to create powerful, cost-efficient, innovative AI-based products, while staying within the systems’ thermal constraints.”

Image: Hailo