Online tool for fast AI optimization and benchmark

STM32Cube.AI Developer Cloud is a free-of-charge online platform and services allowing the creation, optimization, benchmarking, and generation of AI for the STM32 microcontrollers. It is based on the STM32Cube.AI core technology.
Benefits include access to online service to benchmark AI on STM32 boards. It comes with the ST Model zoo including reference model, training script and application examples.

Main features

Online GUI secured by ST user credentials
AI optimizer and visualizer
AI benchmark service on various STM32
STM32 model zoo with training scripts Application code examples
Model quantization tool

An easy to use online platform

Create, optimize, and benchmark your neural network for STM32. No software installation, no board required.

Load NN model
Optimize NN model
Load a model from ST model zoo or own your trained model from your favorite AI framework: Tensorflow Lite, Pytorch, ONNX, scikit Learn, … Supported format are FLOAT32 or quantized INT8.
Select your preferred model optimization options between the RAM or latency most critical parameter.
STM32Cube.AI Developer Cloud screen capture
Perform post-training quantization to further reduce your NN model parameter size.
STM32Cube.AI Developer Cloud screen capture
Access the online benchmark service to remotely evaluate your AI performance on various STM32 boards.
STM32Cube.AI Developer Cloud screen capture
Visualize and optionally export the performance benchmark report.
STM32Cube.AI Developer Cloud screen capture
Generate your optimized AI code in C-code or in a format compatible with the STM32CubeMX software suites.
STM32Cube.AI Developer Cloud screen capture

REST APIs – Seamless integration in your online workflow

You can access all the services of the STM32Cube.AI Developer Cloud directly online via a REST (Representational State Transfer) API. This makes it very easy to automate tasks or integrate the tool into your MLOps flow.

STM32 model zoo – Find the best edge AI model

The STM32 AI model zoo is a collection of pre-trained machine learning models that are optimized to run on STM32 microcontrollers. Available on GitHub, this is a valuable resource for anyone looking to add AI capabilities to their STM32-based projects.

– A large collection of application-oriented models ready for re-training
– Scripts to easily retrain any model from user datasets
– Application code examples automatically generated from user AI model

Get started with the Dev Cloud

Discover how to optimize your AI Neural Network and create processing libraries for your STM32 project