Not even that, transistors (at least the ones I know the foundational mechanisms of, like MOSFETs, but I am by no means an expert) can be understood perfectly well with a classical or semi-classical approach to electrodynamics. For more modern (and much less widespread and well developed) things like optical cavities or any implementation of quantum computing, QED starts playing a role, even though in my limited experience, the full blown quantum field theory approach is not needed either. QFT is fundamental physics research with little to no practical application outside of its field of research - yet. But of course, that does not invalidate the research done in the field and certainly not the mathematical foundations of it.
Would be basically be at a pre-transistor stage without QED?
Not even that, transistors (at least the ones I know the foundational mechanisms of, like MOSFETs, but I am by no means an expert) can be understood perfectly well with a classical or semi-classical approach to electrodynamics. For more modern (and much less widespread and well developed) things like optical cavities or any implementation of quantum computing, QED starts playing a role, even though in my limited experience, the full blown quantum field theory approach is not needed either. QFT is fundamental physics research with little to no practical application outside of its field of research - yet. But of course, that does not invalidate the research done in the field and certainly not the mathematical foundations of it.