Different approaches for treating the solvent in biophysical simulations are reviewed. They include explicit atomistic (classical fixed‐charge, polarizable, or ab initio), explicit coarse‐grained (polarizable or not), and implicit approaches (dielectric‐based or empirical). The solvent is usually an aqueous electrolyte solution, but it can also be an aqueous mixture or heterogeneous, as in micelles and lipid bilayers. The treatment of the solvent is tied to that of the solute, with implicit solvation exhibiting the highest versatility. Applications of implicit and coarse‐grained solvent modeling include a wide range of biological processes, such as protein folding, ligand binding, lipid self‐assembly, and transmembrane translocation. The advantages and disadvantages of each approach are discussed and thoughts are offered on the optimal choice of method for different problems.