What's the difference between PyInstaller and auto-py-to-exe?

I am developing a small app for macOS that I want to convert as an executable. For this I use auto-py-to-exe. I always thought it's just a GUI for PyInstaller, but this article states, that "Another advantage is that Auto-py-to-exe creates an executable file that is a built version of the source code rather than the original source code". I always thought the results of these both tools should be the same, shouldn't they?

auto-py-to-exe also gives you the PyInstaller command-line prompt. If I execute this, I get the three outputs (folders build and dist and the .spec file). auto-py-to-exe only gives one folder as output, which I thought it should be the same then the dist folder. Is that true? If auto-py-to-exe uses PyInstaller shouldn't there also be a build folder and a .spec file?

I am a bit confused and I don't find a proper manual for auto-py-to-exe explaining this. Maybe somebody can shed some light here :).

auto-py-to-exe is just a UI layer on top of PyInstaller - it calls PyInstaller internally.

Directly from auto-py-to-exe's README.md

A .py to .exe converter using a simple graphical interface and PyInstaller in Python.

auto-py-to-exe offers the following on top of PyInstaller:

• A UI to use PyInstaller
• Basic validation on PyInstaller inputs (like paths and .ico file types)
• Builds your script in a temporary directory so there is no need for manual cleanup
• PyInstaller and Python version checks (where some versions can be incompatible)
• Minor default differences where issues typically occur (at the moment we just change the recursion limit)
• Some non-English language support

Disclosure: I am the creator of auto-py-to-exe