We don't know. Pliny says they're the languages of the twenty-two peoples over which he ruled, but our knowledge of who was living in Anatolia under Mithridates and what languages (or dialects) they spoke is pretty spotty. As, mind you, is our knowledge of what exactly our authors mean when they say this. There have been arguments that if it's not an exaggeration the texts really mean dialects, or at the very least languages that must have been closely related to each other, although I don't entirely see how that makes any of this more explicable. Pliny and Valerius Maximus both give the figure at twenty-two languages--Aulus Gellius reports it as twenty-five