Hydrological modeling tools have progressively advanced their functionalities, which has facilitated the implementation of distributed hydrological modeling studies in urban watersheds. Gridded, two-dimensional (2D) hydrologic and hydraulic modeling tools have been applied to both natural and built environments to determine flood extents, high-velocity regions, and other relevant results. Distributed hydrological models with Rain-on-Grid (RoG) capabilities have been proposed, where rainfall inputs are directly added to computational cells. This work compares two such models, both from the US Army Corps of Engineers Hydrologic Engineering Center: HEC-HMS (Hydrologic Modeling System) and HEC-RAS (River Analysis System). These models have been used in hydrological studies, though they vary in terms of approaches to consider rainfall abstractions and overland flow routing. Results from these models are compared against hydrological data collected in a semi-urbanized watershed in Lee County, Alabama. Both models were run on the same unstructured mesh using identical rainfall, terrain, land‑use, and soil data. Model simulations are compared over an extended period to evaluate simulated depth, velocity, and flow hydrographs against field observations. The comparison reveals significant differences among the models in terms of water balance, primarily due to their varying representations of shallow groundwater and base flows. Additionally, HEC-HMS demonstrated a generally better performance in the extended period simulations conducted in this study.