Deep eutectic solvents (DES) are room-temperature liquid mixtures constituted of a hydrogen-bonding acceptor (HBA) and a hydrogen-bonding donor (HBD). They have high practical potential due to their versatility, quick preparation, and wide applications. Therefore, it is appropriate to have robust models to predict their properties. In this work, the density gradient theory has been combined with the perturbed-chain statistical associating fluid theory to model and understand the interfacial behavior in systems of deep eutectic solvents. DESs were modeled as mixtures of their constituents, and a methodology is proposed for estimating the chemical potential of DESs to extend their study to the interfacial properties. Available experimental data of hydrophilic and hydrophobic DESs were used to calculate the influence parameters, providing a way to linearize them in terms of the molecular parameters of HBDs and their molar ratio between HBD and HBA. This treatment has made it feasible to predict the thermal dependence of surface tension in most of the DESs analyzed with an average absolute relative deviation of 1.26%. Furthermore, density gradient theory and perturbed-chain statistical associating fluid theory were applied to predict the vapor–liquid surface tension in mixtures of organic compounds with DES. In particular, we have calculated the surface tension in mixtures of ChCl-glycerol and ChCl-lactic acid with water, ethanol, propanol, phenol, acetone, and ethyl acetate without fitting binary interaction parameters. The behavior of density profiles suggests that the surface is enriched with DES components for the DES + water mixtures. In contrast, it is enriched with diluent for the other ternary systems (ethanol, isopropanol, phenol, acetone, and ethyl acetate).