Most of the world’s large molybdenum (Mo) deposits are genetically related to magmas that underwent significant fractional crystallization and fluid exsolution, and the residual silicate melts are commonly peralkaline. To understand the relationship between melt compositions and the genesis of porphyry Mo mineralization, the partition coefficients of molybdenum (DMo) between fluid and felsic melt were determined at 850 °C, 900 °C, and 100 MPa with various values of A/NK [molar Al2O3/ (Na2O + K2O) in the melt] under an oxygen fugacity around the Ni–NiO (NNO) buffer. The results show that DMo decreases from 4.23 to 0.4 with increasing A/NK from 0.75 to 1.36 in the carbon dioxide series (fluid composed of = 0.096 [molar CO2/ (CO2 + H2O)] and 17.5 wt.% NaCl). A similar trend was observed in pure water, with A/NK varying from 0.72 to 1.23, DMo decreases from 0.54 to 0.15. Significantly, DMo in peralkaline melt is consistently higher than that in peraluminous melt, even with the same NBO/T (the ratio of non-bridge oxygen and tetrahedron ion). Our study therefore reveals that Mo is easily distributed into fluids that exsolved from peralkaline magmas. This finding explains why highly fractionated magmas are critical for porphyry Mo mineralization.