One of the major Python bottlenecks is the Global Interpreter Lock that prevent running multicore Python code.  There are several ways to leverage that (QML, C++ code, asyncio, worker process). But it is a huge concern when you decide to implement an application with Python and Qt. The second one is the slower startup time. However, I think the JIT topic is less critical, since we can speed up a critical path in C++.

As far I understand, the GIL is now a real bottleneck for the Python data science community. Thus, things are moving. And starting with Python 3.13, we are now able to build an experimental CPython interpreter with the GIL disabled and then run true multicore Python code.
See What’s New In Python 3.13 and PEP 703 – Making the Global Interpreter Lock Optional in CPython

It would be great to start to document how PySide is impacted by this new feature.

Notice, the relevant Free Threading Howto Guide is not yet available.