cancel
Showing results for 
Search instead for 
Did you mean: 

About ML and AI Development Environment. What is the difference between X-CUBE-AI and NanoEdge AI Studio.

Kenji1
Senior

Hi Team

I want to know the difference between X-CUBE-AI and NanoEdge AI Studio for ML and AI development environment.

  • Can I do the same thing with either dev environment?
  • Are there any restrictions on the devices used?
1 ACCEPTED SOLUTION

Accepted Solutions

Hi Kenji,

To complete Peter BENSCH answer, both tools are complementary:

X-CUBE-AI is used to convert pre trained neural networks into an optimized code for STM32 microcontroller. Which means that, beforehand, you must create a neural network that solves your problem, yourself (with Pytorch or TensorFlow for example)

NanoEdge AI Studio in the other hand, generate ML solution from scratch by benchmarking various model and selecting the best Machine Learning library given your data (it is an autoML software). NanoEdge will then generate C libraries that you can implement on your embedded code for your STM32 MCU.

NanoEdge can achieve good results with small models, small datasets and has low RAM/FLASH usage. X-CUBE-AI can be used for bigger models such as CNN for computer Vision. So given your use case and your state of progress you might choose to use one solution or the other.

You can find much more information here:

Artificial intelligence ecosystem for STM32 - STMicroelectronics

Here is the link to the documentation of all AI solutions: https://wiki.st.com/stm32mcu/wiki/Category:Artificial_Intelligence

View solution in original post

4 REPLIES 4
Peter BENSCH
ST Employee

X-CUBE-AI was developed by STMicroelectronics and is a free-to-use machine learning tool with good capabilities.

NanoEdge AI Studio was developed by Cartesiam, which was acquired by STMicroelectronics in 2021, and can unleash the full power of machine learning. It can be tested free of charge for 3 months and then requires an annual licence (see DB4564).

Both use very different approaches, which is why two separate tools are currently necessary.

Does it answer your question?

Regards

/Peter

In order to give better visibility on the answered topics, please click on Accept as Solution on the reply which solved your issue or answered your question.
Kenji1
Senior

Hi @Peter BENSCH​ 

Yes, thank you!

Hi Kenji,

To complete Peter BENSCH answer, both tools are complementary:

X-CUBE-AI is used to convert pre trained neural networks into an optimized code for STM32 microcontroller. Which means that, beforehand, you must create a neural network that solves your problem, yourself (with Pytorch or TensorFlow for example)

NanoEdge AI Studio in the other hand, generate ML solution from scratch by benchmarking various model and selecting the best Machine Learning library given your data (it is an autoML software). NanoEdge will then generate C libraries that you can implement on your embedded code for your STM32 MCU.

NanoEdge can achieve good results with small models, small datasets and has low RAM/FLASH usage. X-CUBE-AI can be used for bigger models such as CNN for computer Vision. So given your use case and your state of progress you might choose to use one solution or the other.

You can find much more information here:

Artificial intelligence ecosystem for STM32 - STMicroelectronics

Here is the link to the documentation of all AI solutions: https://wiki.st.com/stm32mcu/wiki/Category:Artificial_Intelligence

Hi @Julian E.​ -san

Thank you for the information.