Artificial intelligence chip firm Hailo Technologies Ltd. mentioned in the present day it’s launching two new acceleration modules that can enhance the processing capabilities of edge gadgets that run its specialist {hardware}.

Hailo burst onto the AI scene in 2019 with a custom-made processor for operating deep learning workloads at the edge of the community. The firm, which is primarily targeted on the automotive sector, mentioned at the time that its Hailo-8 Deep Learning chip permits gadgets resembling autonomous automobiles, good cameras, drones and AR/VR platforms to run refined deep learning functions at the edge that would beforehand be hosted solely in cloud knowledge facilities.

The Hailo-8 processor, which is smaller than a penny, was constructed from the floor up with fully redesigned reminiscence, management and compute structure parts that allow “higher performance, lower power and minimal latency.” Hailo additionally gives a software program growth equipment for builders to construct apps custom-made for the {hardware}.

Hailo is well-funded, having raised $60 million in its most up-to-date Series B funding spherical in May.

The firm mentioned its new M.2 and Mini PCIe high-performance AI acceleration modules help in style deep learning frameworks resembling Tensorflow and the Open Neural Network Exchange. They may be built-in with its Hailo-8 processor to speed up quite a few varieties of deep learning functions, together with picture classification, object detection, segmentation, pose estimation and extra.

Hailo makes some massive claims about the new modules’ capabilities. It gives the instance of “fanless AI edge boxes,” that are used to attach quite a few cameras or sensors to a single clever processing machine at the edge. It mentioned that the modules allow the Hailo-8 processor to ship as much as 26 tera-operations per second, with an influence effectivity of three TOPS/W.

Hailo mentioned the module-accelerated Hailo-8 chip compares nicely with rival processors on a number of widespread neural community benchmark assessments. For instance, it achieved a 26-times larger frames per second fee than Intel’s Myriad-X1 modules, and a 13-times larger fee than Google’s Edge Tensor Processing Unit modules.


Hailo Chief Executive Orr Danon mentioned the new modules will show to be a recreation changer for gadgets at the edge.

“Manufacturers across industries understand how crucial it is to integrate AI capabilities into their edge devices,” Danon mentioned. “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

Since you’re right here …

Show your help for our mission with our one-click subscription to our YouTube channel (under). The extra subscribers we’ve got, the extra YouTube will counsel related enterprise and rising expertise content material to you. Thanks!

Support our mission:    >>>>>>  SUBSCRIBE NOW >>>>>>  to our YouTube channel.

… We’d additionally wish to let you know about our mission and how one can assist us fulfill it. SiliconANGLE Media Inc.’s enterprise mannequin is predicated on the intrinsic worth of the content material, not promoting. Unlike many on-line publications, we don’t have a paywall or run banner promoting, as a result of we wish to maintain our journalism open, with out affect or the have to chase site visitors.The journalism, reporting and commentary on SiliconANGLE — together with stay, unscripted video from our Silicon Valley studio and globe-trotting video groups at theCUBE — take loads of arduous work, money and time. Keeping the high quality excessive requires the help of sponsors who’re aligned with our imaginative and prescient of ad-free journalism content material.

If you want the reporting, video interviews and different ad-free content material right here, please take a second to take a look at a pattern of the video content material supported by our sponsors, tweet your help, and maintain coming again to SiliconANGLE.