Solubility is one of the most important physicochemical properties of a drug, since it is the basis for any technological development, as well as one of the most important information when carrying out environmental evaluations. In this work, the Apelblat, Buchowski-Ksiazczak, van't Hoff-Yaws, NRTL and Wilson models were evaluated in the prediction of the solubility of three drugs of environmental interest because they are considered hazardous xenobiotics for the environment by the Norman network. According to the evaluation of the quality of the adjustment by calculating the mean absolute percentage error(MAPE), it was established that all the models allow calculating the solubility of the three bioactive substances with very low error percentages, i.e. Apelblat 2.98%, van't Hoff-Yaws 2.97%, Buchowski-Ksiazczak 3.35%, NRTL 7.37% and Wilson is 10.88%, which indicates that the models can be an important tool for optimizing processes involving the solubility of these drugs.