A pseudo validation algorithm, which is capable of identifying the prediction uncertainty through recognizing and quantifying the different uncertainty sources in a hydrologic model, is manipulated as an instrument for hydrological model reliability assessment. For implementation, the pseudo validation algorithm is manipulated in order to compare TOPMODEL with different vertical flux calculation components, which have been applied to two Japanese basins. An index, which originates from the Nash-Sutcliffe efficiency, named Model Structure Indicating Index (MSII), is used to quantify the model reliability under different magnitudes of input uncertainty. The results show that within a small magnitude of input uncertainty, the reliability of a five parameter TOPMODEL is worse than a six parameter TOPMODEL. However, within larger magnitudes of the input uncertainty, the reliability of the five parameter TOPMODEL is better than that of the six parameter TOPMODEL, this shows that the pseudo validation algorithm can be used as a reference for hydrological modeling.