With four times the performance and 20% more power efficiency than its predecessor, and the ability to scale from 128 to 2048 MAC units (4 TOPs of arithmetic at 1GHz), the Ethos-U85 is able to power applications such as factory automation and commercial or smart home cameras that require higher performance. Ethos-U85 offers the same consistent toolchain, so partners can leverage existing investments for a seamless developer experience. What's more, the new Ethos-U85 NPU supports AI frameworks such as TensorFlow Lite and PyTorch.

The Ethos-U85 supports the Transformer architecture and Convolutional Neural Networks (CNNs) for AI inference, which will drive new applications, especially for vision and generative AI use cases, and is effective for tasks such as understanding video, filling in missing parts of an image, or analysing data from multiple cameras for image classification and target detection. .

As microprocessors are deployed in more high-performance IoT systems such as industrial machine vision, wearables, and consumer robotics, the Ethos-U85 is designed to work in conjunction with Arm's leading Armv9 Cortex?-A CPU to accelerate processing of Machine Learning (ML) tasks and provide energy-efficient edge inference capabilities for a broader range of higher performance devices.

To date, the Arm Ethos NPU family has more than 20 licensing partners, with Alif Semiconductor and Infineon being early adopters of the new Arm Ethos-U85 NPU.

ML workloads for next-generation edge AI applications require high performance in an energy-efficient manner," said Reza Kazerounian, co-founder and president, Alif Semiconductor, "Alif is the first company in the industry to introduce an edge AI solution based on the Arm Cortex-M55 and Ethos-U55. Alif was the first in the industry to introduce a solution based on the Arm Cortex-M55 and Ethos-U55 for edge AI, and we are very excited and looking forward to the reintroduction of Arm's new AI technology, the Ethos-U85 NPU, which will bring the compute performance needed for our new generation of Ensemble family microcontrollers and converged processors to enable future edge AI and vision use cases."

Steve Tateosian, senior vice president of Infineon's Industrial MCU, IoT, Wireless and Computing business, said, "Edge AI use cases are becoming increasingly complex, and secure, high-performance compute systems are essential to capture the opportunities of the AI era. We look forward to building on our longstanding partnership with Arm to enable these complex systems through the Arm Ethos-U85 and its Transformer architecture support for edge AI deployments."

Corstone-320: New IoT Reference Design Platform Accelerates Deployment of Voice, Audio and Vision Systems

The Arm Corstone-320 IoT reference design platform integrates Arm's highest-performance Cortex-M CPU, the Cortex-M85, Mali?-C55 ISP, and the new Ethos-U85 NPU to deliver the performance needed for a wide range of edge AI applications, such as voice, audio and vision. applications, such as real-time image classification and target recognition, or enabling voice assistants with natural language translation on smart speakers. The reference design platform covers software, tools and support, including Arm virtual hardware. The combined hardware and software nature of the platform will enable developers to start software development before the physical chip is ready, accelerating the product process and reducing time-to-market for increasingly complex edge AI devices.