The Armv9 Edge AI Platform – introduced earlier this year – includes the Arm Cortex-A320 CPU and Arm Ethos-U85 NPU to enable on-device AI models with over one billion parameters, alongside advanced Armv9 security technologies, such as Pointer Authentication Code (PAC), Branch Target Identification (BTI), and Memory Tagging Extension (MTE), to protect critical applications and data at the edge.
Combining Armv9 architectural features, such as Scalable Vector Extension 2 (SVE2), for machine learning (ML) performance and the efficiency of Ethos-U85, delivers the foundation for next-generation edge AI applications, such as smart cameras, smart home applications, industrial automation, and advanced human–machine interfaces, that enable smarter, more natural interactions through vision, voice, and gesture.
This allows machines to perceive and respond like humans, while keeping inference and data processing securely on-device.
Cortex-A320 will be available through Arm Flexible Access in November 2025, with Ethos-U85 to follow in early 2026.
Arm Flexible Access gives partners up-front, low-cost or no-cost (for qualifying startups) access to a wide range of Arm technology, tools, and resources. The model has proven to be a catalyst for innovation, allowing companies to experiment and iterate designs freely, only paying license fees for the technology they use in final designs.
Over the past five years, around 400 successful tape-outs have been achieved through Arm Flexible Access, with more than 300 active members participating.
More than half of these designs target embedded and IoT applications, bringing new intelligence to the edge.
