Arm, a big semiconductor architecture company, is jumping onto the AI train with new edge AI tech built into its latest chips.
Today, the Cambridge, England-based company unveiled its Ethos-U85 Neural Processing Unit (NPU) and the Corstone-320 IoT Reference Design Platform for edge AI applications. The idea is to add more brains to the internet of things.
Arm hopes these foundational chip technologies will revolutionize edge AI applications, offering performance, efficiency, and ease of integration.
Ethos-U85 NPU: Unleashing High-Performance Edge AI
Arm's latest chip design is the Ethos-U85.
The Ethos-U85 neural processing unit represents Arm's most powerful and efficient NPU to date, catering to the escalating demands of edge AI applications.
Boasting a four-times performance boost and 20% higher power efficiency compared to its predecessor, the Ethos-U85 is engineered to excel in scenarios such as factory automation and smart home cameras.
With a scalable architecture ranging from 128 to 2048 MAC units, equivalent to 4 TOPs @1GHz, the Ethos-U85 delivers exceptional performance while maintaining energy efficiency. Its support for AI frameworks like TensorFlow Lite and PyTorch ensures seamless integration for developers, while its compatibility with Transformer Networks and Convolutional Neural Networks (CNNs) enables diverse AI use cases, including vision and generative AI tasks.
The Ethos family of NPUs has already garnered significant industry interest, with over 20 partners licensing the technology. Early adopters of the Ethos-U85, such as Alif and Infineon, attest to its potential to drive innovation in edge AI solutions.
“Machine learning workloads for the next generation of edge AI applications demand high performance in a power efficient manner,” said Reza Kazerounian, president of Alif Semiconductor, in a statement. “Alif was the first to market with an edge AI solution based on Arm Cortex-M55 and Ethos-U55, and we welcome Arm’s latest AI technology, Ethos-U85, 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.”
Corstone-320 IoT reference design platform
Corstone IoT reference design from Arm.
In tandem with the Ethos-U85 launch, Arm introduces the Corstone-320 IoT Reference Design Platform, a solution designed to accelerate the development of edge AI systems. Combining the power of the Arm Cortex-M85 CPU, Mali-C55 Image Signal Processor, and Ethos-U85 NPU, the Corstone-320 platform enables real-time processing of voice, audio, and vision data.
This integrated hardware and software platform facilitates rapid prototyping and development of edge AI applications, offering tools, support, and Arm Virtual Hardware for streamlined deployment. By enabling software development ahead of silicon availability, the Corstone-320 platform significantly reduces time-to-market for complex edge AI devices.
“Edge AI use cases are becoming increasingly sophisticated and require secure, high performance compute systems to deliver on the opportunities of the AI era,” said Steve Tateosian, senior vice president at Infineon, an Arm licensee, in a statement, “We look forward to building on Infineon’s long-standing partnership with Arm and enabling these sophisticated systems with Arm Ethos-U85 and the transformer network support it provides for edge AI deployments.”
Arm designs the Arm architecture, which is used as the foundation for chips created by 15 million developers around the world. When Arm releases designs such as those being announced today, the licensed partners can design and manufacture their own chips based on the designs.