There is no relativistic QM beyond the quasi-nonrelativistic approximations.
QFT is the most general framework for all kinds of quantum theory. You can describe usual non-relativistic quantum mechanics also in terms of a non-relativistic QFT. It's known as "second quantization", because formally you get it by "quantizing the Schrödinger field", but that's a misnomer, because it's just the same non-relativistic quantum mechanics just expressed in a different way.
Now, since QFT is a more general framework, there are physical situations, where you can use QFT but not QM. That's always the case if you don't deal with a fixed conserved number of particles but with creation and annihilation processes, and that's the case in relativistic quantum theory, because you always there's some chance to create and destroy particles in scattering processes at relativistic energies.
Even in non-relativistic theory you can have such cases, if it comes to the description of condensed matter. There you often can describe complicated phenomena by socalled quasi-particles. This technique was ingeniously discovered by Landau. There you describe excitations of a condensed-matter system (usually assumed to be close to thermal equilibrium) by quasiparticles. These are not real particles, but the math of the usually non-relativistic QFT looks right the same. An example are the oscillations of a solid, which are nothing else than sound waves. These you can describe by quasiparticles which are rightfully named "phonons". These can be destroyed and created and thus must be described by a (in this case non-relativistic) QFT.