Ritonavir (RTV) is a weakly basic drug with a pH-dependent solubility. In vitro dissolution and supersatuation behaviors of three Norvir oral products including the tablet, powder, and solution were investigated by two biorelevant dissolution methods with pH alteration: a two-stage dissolution test and a biphasic dissolution-partition test. The two-stage dissolution test revealed a high degree of supersaturation of RTV from these products accompanied by the occurrence of liquid-liquid phase separation (LLPS) in biorelevant dissolution media. Higher, stable apparent RTV concentrations were observed in the FeSSIF-V2 as compared to those in the FaSSIF-V2, which suggested a food effect with higher exposure in the fed state. This is inconsistent with the evaluation in vivo. The biphasic test revealed significantly lower degrees of supersaturation of RTV in the aqueous media from these dosage forms as compared to results of the two-stage dissolution test. RTV concentrations in octanol at 6 h obtained from the tablet and powder with the use of the biorelevant media are consistent with corresponding in vivo AUC and Cmax under the fasting and moderate fat fed (MFF) states and predict the food effect. The underlying mechanisms responsible for the food effect are also proposed. Fractional partition profiles of RTV obtained in octanol from these three Norvir oral products are in agreement with the corresponding fractional absorption profiles in vivo under both the fasting and MFF states. This study reveals a complex interplay among the dissolution, precipitation, and partition processes from these formulations that dictate the oral exposure of RTV.