Arm accelerates Edge AI with Ethos-U NPU and IoT reference design platform

2 mins read

With edge AI scaling rapidly, silicon developers are having to navigate growing system and software complexity, an ever-increasing demand for AI performance and pressure to accelerate their time-to-market.

Credit: Planetz

At the same time, software developers need more consistent, streamlined experiences and easy integration with emerging AI frameworks and libraries.

In response, Arm has announced the Arm Ethos-U85 which is capable of delivering 4x the performance uplift and 20% higher power efficiency compared to its predecessor while scaling from 128 to 2048 MAC units (4 TOPs @1GHz).

“The Ethos-U85 is addressing applications where we see even greater performance demands such as factory automation and commercial or smart home cameras,” said Paul Williamson, SVP and GM of the IoT LoB, Arm. “It offers the same consistent toolchain so partners can leverage existing investments for a seamless developer experience. Importantly, it provides support for AI frameworks such as TensorFlow Lite and PyTorch.”

The Ethos-U85 supports Transformer Networks as well as Convolutional Neural Networks (CNNs) for AI inference. Transformer Networks will drive new applications, particularly in vision and generative AI use cases for tasks like understanding videos, filling in missing parts of images or analysing data from multiple cameras for image classification and object detection.

“With the deployment of microprocessors into more high-performance IoT systems for use cases such as industrial machine vision, wearables and consumer robotics, we’ve designed the Ethos-U85 to work with our leading Armv9 Cortex-A CPUs, to accelerate ML tasks and bring power-efficient edge inference into a broader range of higher-performing devices,” explained Williamson.

The Ethos family of NPUs has been licensed by more than 20 partners to date, and early adopters of the new Ethos-U85 include Alif and Infineon.

“Machine learning workloads for the next generation of edge AI applications demand high performance in a power efficient manner,” said Reza Kazerounian, co-founder and president, Alif Semiconductor. “Alif was the first to market with an edge AI solution based on Arm Cortex-M55 and Ethos-U55, which will deliver the compute performance required for our next generation Ensemble family of microcontrollers and fusion processors to address future edge AI and vision use cases.”

Arm has also introduced the Corstone-320 IoT Reference Design Platform, bringing together its Arm Cortex-M85 CPU, Mali-C55 Image Signal Processor and the Ethos-U85 NPU to support the broad range of edge AI applications for voice, audio, and vision, such as real time image classification and object recognition, or enabling voice assistants with natural language translation on smart speakers.

The platform includes software, tools, and support including Arm Virtual Hardware.

“This combination of hardware and software will help to accelerate product timelines by enabling software development to start ahead of silicon being available, rapidly improving time to market for these increasingly complex edge AI devices,” said Williamson.