STM32Cube AI Studio is now available as ST’s new standalone environment for optimizing, validating, and deploying AI models on STM32 MCUs.
Evolving from the X-CUBE-AI legacy, it brings a unified workflow, improved performance evaluation, and a modern user experience designed for both embedded engineers and AI developers.
What’s new for developers
A stronger foundation for embedded AI
- Built on the trusted X-CUBE-AI legacy to ensure consistent, reliable embedded AI performance
- Based on ST Edge AI Core technology for aligned performance across the STM32 AI toolchain
- Separates embedded and AI tasks to reduce project complexity
- Modern, intuitive UI for smoother, faster development
New capabilities for embedded model optimization, validation & acceleration
- Detailed model performance reports, including latency and memory usage
- Experiment comparison tools to understand design trade‑offs
- Accuracy validation on both host and STM32 target devices
- Layer-level inference metrics visualization for deeper optimization
- Optimized model compilation for STM32 devices with the Neural‑ART Accelerator
Smooth integration within the STM32Cube ecosystem
- Fully compatible with STM32Cube tools
- Reuses existing STM32 projects to simplify upgrades and transitions
- Generates code and templates compliant with STM32 development guidelines
- Provides secure unified authentication and API access for seamless integration with other STM32Cube tools
Additional resources
- Download it here
- Watch this tutorial and get started quickly
- Read the wiki articles
First published on Feb 18, 2026
15 Comments
