Arm, a leading compute platform firm, has introduced the groundbreaking Armv9 edge AI platform, specifically tailored for IoT and edge AI systems-on-chip (SoC) to enhance machine learning capabilities. This innovative platform offers cost-effective solutions with easy access to advanced AI performance and security features for edge devices through Arm Flexible Access.
Featuring the Arm Cortex-A320 CPU and Arm Ethos-U85 NPU, this platform supports on-device AI models with over one billion parameters. Additionally, it incorporates advanced security measures like Pointer Authentication Code (PAC), Branch Target Identification (BTI), and Memory Tagging Extension (MTE) to ensure robust data protection.
Arm unveils the Armv9 edge AI platform, revolutionizing IoT and edge AI with enhanced machine learning capabilities and advanced security features through Arm Flexible Access.
This cutting-edge platform is specifically designed for next-generation edge AI applications such as smart cameras, robotics, and smart home products.
Arm Flexible Access streamlines innovation and time-to-market for startups and OEMs by lowering entry barriers. The Cortex-A320 is set to launch in November 2025, followed by the Ethos-U85 in Q1 2026.
Arm Flexible Access has facilitated over 400 tape-outs, with up to 300 active members at any given time, encompassing startups and major OEMs. More than half of the workloads on Arm Flexible Access target embedded and IoT applications.
Armv9 architecture combines machine learning prowess with performance enhancements from SVE2 and efficiency optimizations from Ethos-U85. This platform aims to enhance human-machine interactions in vision, voice, and gesture recognition applications.
Arm is committed to democratizing advanced technology and computing access within its ecosystem. By 2028, AI is expected to play a central role in initiating new IoT projects leveraging Arm technology.
Through Arm Flexible Access, companies can access a wide range of Arm technologies and tools at low or no cost (for eligible startups), with licensing fees applicable only for the technologies utilized in final designs, fostering innovation and experimentation.
Leading partners like Raspberry Pi, Weeteq, Hailo, and SiMa.ai have spearheaded the development of cutting-edge AI technologies leveraging Arm Flexible Access.
Arm is actively expanding its ecosystem to meet the growing demand for performance, efficiency, and security in edge AI applications.
Related
ARM | Arm flexible access | armv9 | edge AI | hardware | IoT
