This study tests the overall consistency of geothermobarometric calibrations and thermodynamic internally consistent databases commonly used to retrieve temperature and pressure conditions in metapelitic and igneous mafic protolith metamorphosed at granulite-facies conditions. We examine a lithological sequence characterized by interbedded mafic and metapelitic protoliths that underwent partial melting at the same time during peak T conditions. Textural equilibrium and the lack of complex chemical zoning in rock-forming minerals from the two contrasting protoliths permit the application of commonly utilized ion-exchange and net transfer thermobarometers. By combining garnet–biotite and garnet–cordierite thermometers with the well-known net-transfer barometers, and using thermodynamic properties from both Berman's and Holland and Powell's databases, we demonstrate that temperature and pressure estimated for the metapelitic migmatite are independent of the chosen internally consistent thermodynamic database. Peak pressure–temperature (P–T) estimates in the metapelitic mineral assemblage are also consistent with those inferred from pseudosection models constructed with the whole-rock composition of the metasedimentary migmatite. The most tenable values for the peak temperature–pressure condition retrieved in the metapelitic migmatite are taken to provide the best constraint on which to test thermobarometric results from the mafic migmatites. Two-pyroxene thermometry using the database from Berman (1988) and the amphibole–plagioclase thermometer with the calibration by Holland and Blundy (1994) yields values broadly consistent with those determined in the metapelitic migmatite. By contrast, other calibrations of the two-pyroxene thermometer tend to overestimate temperature by at least several tens of degrees. The construction of a P–T pseudosection for the mafic migmatite composition provides further evidence for both selecting the most accurate thermometer and determining the peak pressure of a mafic migmatite.