Your searches are likely confused by the tendency for Rosetta to call gdt_ts “gdt_mm” instead, for strange historical reasons.
The functions you want are in the core.scoring namespace. core.scoring.CA_gdtmm( pose1, pose2 ) is likely the most straightforward way of calculating, (this use the average of the 1/1, 2/2, 3/3, 3/4, and 4/7 rmstol/disttol maxsub metrics) but there’s parameter variants to include/exclude certain residues, as well as gdtha(), plus CA_maxsub() and even xyz_maxsub() functions if you want to roll your own GDT-like metric.
.
