Deep eutectic solvents (DESs) have attracted significant interest in recent decades due to their environmentally friendly nature and wide range of industrial applications. This chapter explores the use of in silico Quantitative Structure-Property Relationship (QSPR) modeling to predict and optimize DES properties of interest. The focus will be mainly paid to non-biological properties, reviewing methodologies for data collection, descriptor calculation, model generation, evaluation, and interpretation. The chapter highlights the frequent use of conventional 2D and 3D descriptors, quantum-chemical COSMO-RS descriptors, and thermodynamic properties in QSPR modeling. Furthermore, it emphasizes the surge in research interest driven by advancements in machine learning (ML) techniques. Techniques like multilayer perceptron (MLP) neural networks, random forests (RF), and support vector machines (SVM) are now commonplace for model development. Looking ahead, the focus should shift more towards multi-objective optimization techniques to address several DES properties simultaneously and achieve a better-tailored design for specific applications. Additionally, the importance of setting up and maintaining online DES databases for facilitating QSPR model development with larger datasets is discussed. Finally, the chapter explores the potential of novel cheminformatics tools, such as transformer-based QSPR modeling, for further advancing research in this area.