Extreme El Niño events entail important socio-economic challenges, both in regions such as South America directly affected by their impacts and in regions around the world that are influenced by the associated teleconnections. Uncertainty remains about the ability of recent climate models to reproduce the characteristics and impacts of extreme El Niño events. In this study, we evaluate the ability of 32 CMIP6 models to simulate extreme El Niño events, focusing on their occurrence, their seasonal evolution, and the characteristics of the associated atmospheric moisture flux divergence. Our results reveal the reasonable performance of the CMIP6 ensemble in reproducing the observed anomalies and seasonal cycles of extreme El Niño events. The ensemble mean also captures the average temporal evolution and magnitudes of moisture flux anomalies, but fails to reproduce some important aspects of the associated spatial patterns. Most individual models have marked deficiencies in adequately simulating the seasonal cycle of atmospheric moisture flux divergence dynamics and reproducing a clear distinction between moderate and extreme events. The latter indicates that the atmospheric–ocean coupling and resulting precipitation anomaly patterns over Ecuador and northern Peru are still not correctly reproduced by the individual models. These deficiencies echo previous studies and underscore the limitations of current global climate models in providing reliable insights into the impacts of climate change on El Niño extremes and their consequences for regional atmospheric dynamics and precipitation. This work highlights the need for further research to improve model representations of extreme El Niño events and their associated impacts on vulnerable regions, thereby facilitating more effective risk management and adaptation measures.