Cortex-A320 CPU and Ethos-U85 NPU Set New Industry Standards
March 11, 2025
News
Recently Arm introduced its Armv9 edge AI platform featuring the new Arm Cortex-A320 CPU and the Arm Ethos-U85 NPU, a leading AI accelerator delivering efficient and secured intelligence at the edge. The solution provides an 8x improvement in machine learning (ML) performance compared to the Cortex-M85-based platform launched last year.
The Cortex-A320 leverages Armv9 architectural features, such as SVE2 for ML performance, with a 10x ML performance uplift and 30 percent scalar performance lift compared to the Cortex-A35.
The platform’s Armv9.2 architecture supports security features like Pointer Authentication (PAC), Branch Target Identification (BTI) and Memory Tagging Extension (MTE) to even the smallest Cortex-A devices while operating in rough environments managing sensitive data.
Delivering AI capabilities and tools for developers, the Cortex-A320 extends the Armv9 architecture to IoT, optimizing power efficiency and ensuring software compatibility. According to the company, with the latest addition, Arm now offers the most advanced family of Armv9 Cortex processors for IoT.
Arm is extending its set of compute libraries for developers of AI frameworks, its Arm Kleidi, to IoT. It is designed to enhance AI and ML workloads on Arm-based CPUs with no additional developer work required.
KleidiAI is integrated into general IoT AI frameworks such as Llama.cpp and ExecuTorch or LiteRT via XNNPACK, increasing the efficiency of key models, including Meta Llama 3 and Phi-3. Per example from Arm, KleidiAI brings up to 70% more performance to the new Cortex-A320 when running Microsoft’s Tiny Stories dataset on Llama.cpp.
“The new Arm edge AI platform will enable our customers to run nucleus lite, a lightweight device runtime of AWS IoT Greengrass for constrained edge devices with minimal memory needs, on Armv9 technology. This seamless integration between the two technologies provides an optimized solution for developers to build modern edge AI applications like anomaly detection in precision agriculture, smart manufacturing, and autonomous vehicles,” commented Yasser Alsaied, Vice President of loT, AWS
For more information, visit arm.com/.
Chad Cox. Production Editor, Embedded Computing Design, has responsibilities that include handling the news cycle, newsletters, social media, and advertising. Chad graduated from the University of Cincinnati with a B.A. in Cultural and Analytical Literature.
