Cell+atom relaxation with  scph without 4th order FC

Hi,

Is it possible to run self-consistent phonon calculations up to 3rd order FCs when MODE=SCPH and RELAX_STR = 2 in anphon, i.e., relaxing cell size and atoms? The 4th order calculations require a substantial number of displacements and I am hoping to avoid it if possible.

Thanks,
Vahid