Arm today announced the new Cortex-M42 processor, expanding its portfolio to bring efficient AI compute capabilities to more intelligent devices. Evolving Cortex-M-based designs to bring advanced AI capabilities to endpoints, Cortex-M42 combines Arm Helium technology with a compact AHB-based design, making it easier for silicon partners to upgrade existing Cortex-M4 and Cortex-M33-based microcontrollers (MCUs). This supports a wide range of cost- and area-sensitive applications, from audio processing and keyword spotting in wearables to sensor analysis, condition monitoring and signal processing in connectivity and industrial controllers.
These capabilities are also relevant to physical AI, where intelligent machines rely on distributed embedded processing for sensing, anomaly detection, predictive maintenance and control. For example, Cortex-M42 could enable local signal processing and condition monitoring of robotic motors and joints, helping identify abnormal behavior or signs of equipment wear without performing safety-critical motion-control functions. Built on the Armv8.1-M architecture, the processor delivers up to 3x the vector compute capability and 2x the vector power efficiency of Cortex-M33, while Arm Custom Instructions provide additional flexibility to optimize processors for specific workloads.
The addition builds on the continued expansion of the Arm Flexible Access portfolio, including Arm Cortex-M52, Arm Cortex-M55, Arm Cortex-M85, Arm Ethos-U85 NPU, Arm Corstone-320 reference design platform and the Armv9 Edge AI platform, providing partners with a broader range of compute options for intelligent devices. As of November 2026, Arm Flexible Access has 319 active members and has enabled more than 500 unique tape-outs. Arm Total Access continues to grow, with 60 partners to date and more than 600 projects undertaken through the program since launch.
Source: TechPowerUp News · Summarized by HeadlinesBriefing