Skip to main content
stoicLemons
Associate
July 29, 2021
Solved

STM32L1 and STM32L0 for running Neural Networks

  • July 29, 2021
  • 1 reply
  • 4027 views

Can STM32L1 and STM32L0 be used for AI development (is. running inference of neural networks)? I was hoping to use those chips with X-CUBE-AI. Yet, when I read the user manual, those chips were left out of the list of microcontrollers. Why is this so?

User Manual: https://www.st.com/resource/en/user_manual/dm00570145-getting-started-with-xcubeai-expansion-package-for-artificial-intelligence-ai-stmicroelectronics.pdf

Also, there is STM32Cube.AI which I assume will work with STM32L1 and STM32L0. Is this correct?

This topic has been closed for replies.
Best answer by fauvarque.daniel

Our current AI offering is not supported on those series. We only support Cortex-M4 Cortex-M7 and Cortex-M33 based series mainly to benefit from Cortex special instructions to optimize the processing and the available floating point unit.

We have in our roadmap to extend the support to Cortex-M0 and Cortex-M0+ series, the exact delivery date is not yet fully decided

Regards

Daniel

1 reply

fauvarque.daniel
fauvarque.danielBest answer
ST Employee
July 29, 2021

Our current AI offering is not supported on those series. We only support Cortex-M4 Cortex-M7 and Cortex-M33 based series mainly to benefit from Cortex special instructions to optimize the processing and the available floating point unit.

We have in our roadmap to extend the support to Cortex-M0 and Cortex-M0+ series, the exact delivery date is not yet fully decided

Regards

Daniel

stoicLemons
Associate
July 29, 2021

Thank you, that makes sense.