<p>Climate change is expected to intensify precipitation extremes over the Eastern Tibetan Plateau (ETP), as indicated by observational data and model projections. This study investigates the simulation differences in precipitation extremes over the ETP by comparing low-resolution (LR) and high-resolution (HR) models from the High-Resolution Model Intercomparison Project (HighResMIP) protocol in the Coupled Model Intercomparison Project Phase 6 (CMIP6). Results indicate that the variations in precipitation extremes are reproduced well in most HR and LR models over the ETP during the period of 1985 to 2014. Compared with LR models, HR models exhibit significantly lower biases in precipitation extremes over the ETP. Future projections indicate that the intensity of precipitation extremes will increase substantially. We further investigate the underlying physical mechanisms of the increased precipitation extremes using a Physical Scaling Diagnostic approach, suggesting that the dynamic components account for approximately 82% of the total scaling change. These changes in dynamic scaling are the primary drivers of intensified precipitation extremes and model uncertainties. Short-term projections are mainly influenced by the model internal variability, while model uncertainty increases with longer projection periods. These findings highlight the critical influence of dynamic processes on the intensity and uncertainty of precipitation extremes, offering valuable insights into model characteristics and providing decision-makers with more reliable information to support climate mitigation and adaptation strategies over the ETP.</p>