This study conducted and evaluated 44 experiments using the non-hydrostatic version of the regional climate model RegCM4 (RegCM4-NH) and an additional three experiments with RegCM version 5 (RegCM5) over Southeast Asia for the period 2010–2015. The initiative was part of the coordinated regional climate downscaling experiment—Southeast Asia (CORDEX-SEA) project, in preparation for downscaling the latest coupled model intercomparison project Phase 6 (CMIP6) global climate models (GCMs). The RegCM4-NH experiments, forced by the ERA5 reanalysis, were configured using combinations of four cumulus, three planetary boundary layer (PBL), and three explicit moisture schemes. The spatiotemporal variability of simulated 2 m-temperature and rainfall for 2010–2015 was evaluated against observational datasets. The best experiments demonstrated reasonable reproduction of observed annual cycles and spatial distribution, while many exhibited unrealistic biases. A score ranking system was implemented to objectively compare the performance of experiments, enabling the identification of top-ranked experiments for Southeast Asia. The ensemble mean of the 44 RegCM4-NH experiments exhibited commendable performance, ranking 11th overall. Furthermore, the three additional RegCM5 experiments did not yield improved results compared to RegCM4-NH under the same physical configuration, suggesting that opting for RegCM4-NH would be a prudent choice for the CORDEX-SEA community in the forthcoming CMIP6 downscaling cycle for Southeast Asia.