Flooding is a natural disaster and a long-term challenge for urban planning. Numerous studies have stated how flood risk could increase with climate change and urbanization. HEC-RAS and TELEMAC are some of the sophisticated two-dimensional modelling tools to simulate floods. However, they have different numerical schemes, especially their meshing and discretization. A comparison is needed to evaluate how their results can differ. This study simulated a historical flood event in Sapan, a region in the Upper Citarum Watershed, Indonesia, using HEC-RAS and TELEMAC. Additionally, this study investigated how both models simulate flood extent, propagation, velocities, and depths, especially from TELEMAC since it is underutilized in Indonesia. The models’ setup was configured to be equivalent. However, differences remain, especially in the building interpretation. The flood extent from both models was remarkably similar, with a 1.34 % difference in max inundated area. The main differences were velocity distribution around buildings because TELEMAC had a finer mesh. However, with a coarser mesh, HEC-RAS could simulate a similar flood extent through the utilization of subgrid bathymetry information. Both models have demonstrated good performance. HEC-RAS might be more suitable for broad-scale modelling, where inundation depth is the main concern. However, if detailed velocity mapping is needed, TELEMAC might be more suitable.