A top-down reflooding model was developed and implemented into the French best-estimate thermalhydraulic code CATHARE2 V1.3E. A two-dimensional mesh moving along the wall with the quench front is used to resolve the heat conduction equation in the wall near the quench front. The results of the model validation and the first assessment calculations are given. The Winfrith single-tube top-down reflooding experiments were used to validate the model. The influence of wall material, pressure, mass flux, and wall temperature on the quench front velocity are correctly predicted. The REWET-II and PERICLES experiments in rod bundle geometry were used to assess the capabilities of the code to predict simultaneous bottom and topdown rewetting. Comparison of the calculated results with the RE WET II experimental data shows the ability of the new package to calculate the key features of this complex experiment. The existence and progression of two quench fronts in the core are correctly predicted. The maximum cladding temperatures are overpredicted for experiments with combined and upper plenum injection. This difference, which is attributed to a too severe countercurrent flow limit (CCFL) calculated by the code, does not exceed 150°C. With the top-down reflooding option, improved predictions of wall temperatures in the upper part of the core in the PERICLES tests with respect to the previous version of the code were obtained, since this part of the core was rewetted by top-down quenching. To realize further improvements in combined reflooding calculations, the CCFL predicting capabilities of CATHARE should be addressed first. More detailed experimental information and additional data would also be required for in-depth assessment of the models.