I want to learn more about EdgeAI or EmbeddedML, whatever term you like best.

I played with an Arduino Nano, running some of the tutorial projects (motion detection), but I don't think it is up to much.

If anyone here works in this area, what hardware do you use?

  • $\begingroup$ This question is too broad, opinion-based and off-topic. So, it clearly should be closed. Please, see our on-topic page for more info. We focus on the theoretical aspects of AI. This is related to specific hardware that people are using. So, off-topic. $\endgroup$
    – nbro
    Aug 31 at 11:32
  • $\begingroup$ @nbro Apologies, can you point me towards the on-topic page? I cannot seem to locate it. Also, hardware should not be off-topic, as models have to run on something. Just a suggestion: I will aim to keep my questions more to the theoretical aspects going forward. $\endgroup$
    – ofithch79
    Aug 31 at 14:18
  • $\begingroup$ See ai.stackexchange.com/help/on-topic. There's also Data Science SE, where this question may be more suitable or even on-topic. $\endgroup$
    – nbro
    Aug 31 at 15:06
  • $\begingroup$ @nbro Appreciate it, thanks. $\endgroup$
    – ofithch79
    Aug 31 at 19:05

1 Answer 1


8-bit microcontrollers (such as the AVR on the nano) are unlikely to do much in the way of inference. One could in theory enlist/write a matrix library that worked with 8-bit floats but it's unlikely to be particularly efficient.

Modern ARM microcontrollers (such as those on the Nano 32/33) with 32-bit cores and with a floating point unit are much more likely candidates. ARM very helpfully provides the CMSIS NN library to run pre-trained models. Size of model will be limited compared to desktop/server deployed models but some stuff can be successfully run. There are bleeding edge microcontrollers with neural processing units (NPUs) at the time of this writing.

Another option is an FPGA. That is a whole other skill set and the learning curve is steep but they can be much higher performance than a microcontroller.

On the largest side of embedded, application cores with Mali GPUs/etc can do a bit more work. There are USB powered NPUs for processors without a GPU (or powerful enough GPU). We're even beginning to see application cores with built in NPUs.

  • $\begingroup$ Very informative, thanks for the answer. $\endgroup$
    – ofithch79
    Aug 28 at 18:41

Not the answer you're looking for? Browse other questions tagged .