Xavier NX vs AGX for Python NLP tasks real difference

I would use low power device to perform text analysis (NLP) based on Python libraries like transformers or Spacy. What is real difference in comparision NX to larger brother? How faster it is or typical NLP tasks? I don’t know that performance difference justify spend extra money for it.

I think about using prepared models than build on this device. My target is create low power NLP tasks server and using it to run on Jetson data and get result back.

Thank you for your answers.


You can find the comparison for XavierNX and Xavier here: