The recently developed Nonrandom Two-Liquid Segment Activity Coefficient (NRTL-SAC) model [reported by Chen and Song, Ind. Eng. Chem. Res. 2004, 43, 8354] offers a practical thermodynamic framework to predict drug solubility in a wide range of solvents, based on a small initial set of measured solubility data. The model yields satisfactory results in first correlating drug solubility in a few representative pure solvents and then qualitatively predicting drug solubility in other pure solvents. Here, we investigate the applicability of the NRTL-SAC model for predicting drug solubility in mixed solvents for three molecules: acetaminophen, sulfadiazine, and cimetidine. The study shows that the model provides robust correlation and prediction of drug solubility in both pure and mixed solvents, with a qualitative level of accuracy. The model is a useful tool in support of the early stages of crystallization process development and other areas of drug process design.