Flash floods have dramatic economic, natural and social consequences, and efficient adaptation policies are required to reduce these impacts, especially in a context of global warming. This is why it remains essential to develop more efficient flash flood forecasting systems. This study was carried out in order to assess the ability of a new seamless short range ensemble rainfall forecast product, called PIAF-EPS and recently developed by Meteo France, to predict flash floods when it is used as input in an operational hydrological forecasting chain.For this purpose, eight flash flood events that occurred in the French Mediterranean region between 2019 and 2021 were reproduced, using a similar forecasting chain as the one implemented in the French "Vigicrues-Flash" operational flash flood monitoring system. The hydrological forecasts obtained from PIAF-EPS were compared to the hydrological simulations obtained from the radar observations, and to three deterministic forecasts using varied scenarios (future constant rain, deterministic PIAF, and a numerical nowcasting system called AROME-NWC). The verification method applied in this work uses rank diagrams and scores calculated on contingency tables, in an original way. The verification process has been conducted on each 1km² pixel of the territory. The results illustrate the added value of the ensemble approach for flash flood forecasting, and the benefits of the use of a seamless product combining radar observations and numerical nowcasting.