Over the past decades, the unique properties and diverse uses of ionic liquids (ILs) have garnered significant interest from scholars and industry professionals worldwide. The Web of Science search results indicate that several thousand research articles associated with ionic liquids (ILs) are reported annually, demonstrating their importance. ILs are liquefied organic salts made up of organic cations and inorganic or organic anions, remaining liquid at room temperature or below 100 °C. As a result, numerous chemical structures of cation and anion pairs allow researchers to synthesize a limitless variety of salt structures tailored for specific physicochemical properties and applications. While some ILs are toxic and have significant cross-solubility in water, raising concerns about their potential ecosystem and environmental fate, having information on physicochemical properties, or at least an initial estimation, seems necessary. Predictive modeling provides a rational assessment strategy for physicochemical property prediction of ILs established especially on the information of the structures of the ions building them. Also, it allows for the prediction of their properties before synthesis. Computational approaches, commonly recognized as in-silico techniques, can minimize the time, cost, and energy required to perform experimental measurements for a broad range of ILs. The modeling approaches are regressed from empirical property data and can predict the properties of novel ILs that are not utilized in model training. In addition, mathematical models are used in the computer-aided molecular design (CAMD) of ILs with desirable properties. Various approaches have been suggested for developing predictive models for the physicochemical properties of ILs. These methods are grouped into regression (linear or nonlinear) approaches and theoretical (predictive).