The STM32N6 is the latest and most powerful of the STM32 family, and has been optimised for edge-intensive AI algorithms in automotive, smart industry, robotics, drones, healthcare, smart buildings, smart homes, smart farming, and personal electronic applications.
The MCU is the first of the STM32 series featuring ST's proprietary Neural-ART Accelerator, specifically architected for embedded inference, delivering 600 times more machine-learning performance than existing high-end STM32 MCUs.
The STM32N6's Neural-ART Accelerator clocks at 1GHz, providing 600 giga-operations per second (GOPS) at an average of 3 tera-operations per second (TOPS), delivering energy efficiency while allowing machine learning applications requiring an accelerated microprocessor to now run on an MCU.
The device’s machine learning performance makes it possible to run computer vision, audio processing and sound analysis for consumer and industrial applications at the edge, utilising a Cortex-M55 MCU running at 800 MHz, a 4.2 MB embedded RAM that also includes a NeoChrom GPU alongside an H.264 hardware encoder, and a Helium M-Profile Vector Extension.
The STM32N6 delivers a combination of AI performance and flexibility in a small silicon package making is suitable for both embedded systems and wearable devices.
The STM32N6 is supported by the STM32N6570-DK discovery kit, also available to order from Mouser. The kit is a complete demonstration and development platform for the STM32N6 and provides a full range of hardware features to enable users to evaluate the device – it includes USB Type-C, Octoa SPI flash memory and Hexadeca SPI PSRAM devices, Ethernet connectivity, a camera module, and a 5" LCD touchscreen.
Four flexible extension connectors also offer easy expansion capabilities for applications such as wireless connectivity, analogue applications and sensors.
