In the current work, an empirical solvation model based on accessible surface areas was reported, which can be used to predict the solvation free energies for both organic and biological molecules very fast. This solvation model is based on atom-weighted solvent accessible surface area (SAWSA). The parameterization procedure for different kinds of atoms was performed as follows: first, the atoms in a molecule were defined to different atom types based on SMARTS language; then the solvent accessible surface area for each atom (or charged group) was calculated; finally, 4 genetic algorithm (GA) was applied to optimize the solvation parameters for different atom types in order to reproduce the experimental solvation free energies. The derived model possessed promising predictive ability as indicated by the good statistical significance and good prediction on the external test set. Using the solvation model based on all 377 neutral molecules, we have achieved an average unsigned error of 0:51 kcal/mol and standard deviation of 0.46 kcal/mol, which was better than the model proposed by Wang et al. The solvation model developed in the current work was applied to predict the solvation free energies of small organic molecules and proteins. For the,51 small organic molecules, the SAWSA model could give consistent results with the AM1/SM2.1 model in addition to several molecules with large conjugate systems. Moreover, the predictions from SAWSA were much better than those from SM5.0R, a solvation model based on geometry-dependent atom surface tensions. For the IS proteins randomly selected from the Brookhaven PDB database, the solvation free energies predicted by the SAWSA model showed high linear correlation (r = 0.99) than those predicted by PBSA, which were much better than those given by the Ooi model and the Vial model. Finally, we have successfully applied this model to predict the relative binding free energies for four binding modes of EGFR/quinazoline. The most favorable binding mode identified by MM-PBSA could also be correctly recognized by MM-SAWSA. The relative solvation free energies calculated by SAWSA show obvious correlation with those calculated by PBSA. The SAWSA should have potential applications in QSAR, molecular docking, protein folding, free energy calculations, and so forth.