STM32Cube.AI is a free software tool to import and convert pre-trained Machine Learning or Neural Network models into optimized C code for STM32.
Today, embedded hardware devices need to perform more complex AI tasks. To help developers create innovative applications, ST offers a complete ecosystem, including a powerful neural network conversion tool called STM32Cube.AI.
STM32Cube.AI is a tool for users who have prior experience in creating and training deep learning NN models in frameworks such as TensorFlow Lite, Keras, qKeras or Pytorch. STM32Cube.AI allows you to convert pre-trained neural networks into optimized code for STM32 microcontrollers.
Compatible with the entire STM32 portfolio, STM32Cube.AI is an essential tool for anyone looking to integrate neural networks into STM32 designs. This software is accessible via both graphical interface and command line to match various workflows.
STM32Cube.AI is an STM32CubeMX expansion pack (X-CUBE-AI). Seamlessly integrated into the STM32 ecosystem, it is available now for download from STM32CubeMX or www.st.com.
- Allows you to generate a library from pre-trained Neural Networks and typical Machine Learning models, which is optimized for STM32
- Supports the most popular frameworks (Tensor Flow Lite, Keras, qKeras, Pytorch, ONNX and more)
- Easy portability across different STM32 microcontroller series through STM32Cube integration
- Offers the possibility to integrate converted Neural Networks libraries more easily, thanks to application-oriented code examples (function packs)
- Now supports deeply quantized Neural network (BNN, qKeras and Larq format) -> read our blog article
How to test & prototype with STM32Cube.AI
ST platforms include all the connectivity interfaces and displays to evaluate and develop code on target for your use cases.
The STWIN SensorTile wireless industrial node (STEVAL-STWINKT1B) is a development kit and reference design that simplifies the prototyping and testing of advanced industrial IoT applications, such as condition monitoring and predictive maintenance.
The camera module bundle (B-CAMS-OMV) can be used with the STM32 boards featuring a 1 × 30-pin ZIF connector to link multiple cameras and easily implement computer vision on STM32 MCUs. The camera module is compatible with the STM32 discovery kits and evaluation boards that feature a ZIF connector, such as the STM32H747I-DISCO and 32L4R9IDISCOVERY.
This webinar introduces Artificial Intelligence for edge computing and shows you how ST's offer can help you run Neural Networks on microcontrollers and microprocessors.
Thanks to practical examples, you will learn how to run Artificial Neural Networks and how to use STM32Cube.AI to convert them into optimized code for STM32 MCUs.
Use the power of deep learning and hop on board! Discover how ST’s AI solutions, ecosystem, and network of expert partners can support AI application development and help you reduce time to market.