The tropical cyclone genesis potential index (GPI) can reasonably reproduce the position and region of tropical cyclone genesis. It has been widely used as a metric to evaluate the performance of global climate models in simulating tropical cyclone genesis. In this study, the simulation of GPI over the western North Pacific (WNP) by a regional coupled model, FROALS (Flexible Regional Ocean-Atmosphere-Land System) developed by LASG (State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics)/IAP (Institute of Atmospheric Physics), was assessed and compared with a regional atmosphere model, RegCM3. The simulation bias is discussed by analyzing the five variables associated with GPI. The results show that FROALS can reproduce the spatial pattern and seasonal cycle of GPI in the WNP better than RegCM3. Compared with RegCM3, FROALS performs better in modeling the response to ENSO in the interannual variability of GPI in the WNP; this is due to the improved simulation of the intensity and interannual variability of the South China Sea monsoon trough.