The WonderMK K230 vision module is powered by the K230 chip, delivering 13.7× the performance of K210. It integrates a 2MP HD camera, microphone, HD touchscreen, and speaker for accurate visual capture and clear voice interaction. Onboard I2C, UART, and a 10-pin expansion interface allow the module to work with STM32, ESP32, Arduino, Raspberry Pi, and other controllers, as well as various sensors and peripherals.
A built-in user interface supports 1-click AI training, color sampling & recognition, and multi-face recognition, removing complex training workflows and significantly shortening development cycles. Compatible with MicroPython and the official OpenMV libraries, the module enables over 30 AI vision applications, including color recognition, face recognition, human feature recognition, YOLO-based object detection and tracking, code recognition, digit recognition, and self-learning classification.
The WonderMK K230 vision module also supports multimodal large AI models with access to cloud-based LLM APIs such as Alibaba Qwen, allowing seamless switching between text, speech, and vision models. This makes it easy to implement speech recognition, speech synthesis, text recognition, and image understanding, providing robots with enhanced perception, reasoning, and action capabilities for advanced embodied intelligence.

The WonderMK K230 vision model integrates high-performance core hardware, such as an HD camera and a capacitive touchscreen, delivering powerful computing and flexible expansion capabilities, providing reliable support for a wide range of AI applications.