STMicroelectronics Launches the ST Edge AI Suite, Promises Inspiration and Easier Development
New "stepping stone" collects 10 of the company's AI tools, a model zoo, and additional resources including over 50 case studies.
STMicroelectronics has announced the launch of the ST Edge AI Suite, designed to make it easier to build machine learning and artificial intelligence (ML and AI) projects capable of running on edge devices powered by the company's microcontrollers.
"ST Edge AI Suite is the new starting point on st.com for embedded AI development," STMicro's chief innovation officer Alessandro Cremonesi explains. "Here, today's innovators, driven by imagination, can find help to realize tomorrow's smart things that sense, infer, and respond intelligently, autonomously, and efficiently to the world around."
"This environment brings developers simplicity," Cremonesi continues, "through an easy choice of models and data sources, finding the right tools quickly and easily, optimizing and benchmarking, then automatically generating code and libraries — all within one unified framework."
The ST Edge AI Suite collects together a range of software tools and pairs then with a variety of resources, including over 50 case studies to serve as inspiration. There's a model zoo, with support for you to bring your own model instead if preferred, along with a high-performance data logging system and multiple optimization tools to boost performance. The ST Edge AI Developer Cloud provides easy access to benchmarking, the company says, and ready-to-deploy sample code is also available.
Exactly what you'll be able to use depends on what STMicro devices you're targeting: the model zoo, for example, covers the company's STM32 microcontroller and microprocessor families, as well as its sensors with integrated machine learning core (MLC) and intelligent sensor processing unit (ISPU), but not its Stellar automotive microcontrollers that get their own Stellar Studio AI platform. The Edge AI Developer Cloud, meanwhile, covers everything except MLC-equipped sensors — and the NanoEdge AI studio only covers the STM32 microcontroller range and ISPU-equipped sensors.
More information on the ST Edge AI Suite is available on the STMicro website; all tools, the company has confirmed, are free to use.