Liquid-liquid extraction is a potential separation process for the purification and isolation of pharmaceuticals. However, as considerable experimental effort is required to choose an adequate extractant, liquid-liquid extraction is rarely used in the pharmaceutical industry. By applying a thermodynamic model to predict the extraction behavior of pharmaceuticals, the experimental effort required to select a suitable extractant can be decreased substantially. This work demonstrates that PC-SAFT is able to predict the extraction behavior of pharmaceuticals based solely on solid solubility data of the pharmaceutical in pure solvents. Because these data are required for pharmaceutical licensing and registration, they are usually available. To demonstrate the power of the modeling tool, six ternary two-phase systems containing a pharmaceutical intermediate or its impurity were modeled with PC-SAFT. The modeling results for the extraction behavior of the two pharmaceuticals were found to be in good agreement with the experimental data.