Dissolved organic carbon (DOC), primarily sourced from soil organic carbon (SOC), plays a critical role in regional and global carbon cycles. However, the complexities of underlying mechanics and limited observations pose considerable challenges for predictive understanding and modeling of DOC at regional or larger scales, such as model parameterization. Recently, we developed a machine learning-based (ML) map of DOC production rate, which bridges the gap between SOC and DOC leaching flux, allowing simplification of terrestrial DOC representation. Leveraging this advancement, we introduce a DOC module into the riverine component of the Energy Exascale Earth System Model (E3SM), Model for Scale Adaptive River Transport (MOSART), denoted as MOSART-DOC. MOSART-DOC simulates both the transport and transformation of DOC across headwater streams and heavily managed large river networks. MOSART-DOC will enhance our predictive understanding of riverine biogeochemical processes and effectively reduce uncertainties in modeling regional and global carbon cycles in Earth System Models (ESMs). The development of MOSART-DOC paves the way for future research in Earth system modeling for advancing insights into carbon cycling and its profound implications for global environmental change.