The Canadian prairies is a drought-prone region. Because the hydroclimate of the region is influenced by a myriad of regional and large-scale processes, model predictions of Prairie droughts have been problematic. One of the worst droughts that affected the region occurred recently during 1999-2005. In this study, the drought event was simulated with several regional climate models (RCMs) that included different versions of the Canadian Regional Climate Model (CRCM), the Global Environmental Multiscale – Limited-Area Model (GEM-LAM), the Climate version of the Local Model (CLM) and the Regional Spectral Model (RSM). The results are compared with available observations and reanalysis products to assess to what extend aspects of the extreme event could be simulated accurately with state-of-the-art RCMs. In addition, biases in the simulated regional water and energy budgets are related to regional manifestations of known deficiencies in the models where possible. Implications of the results to improving drought predictions for the region will also be discussed.