Warfarin is a widely used anticoagulant, and bleeding complications are the main reason why patients discontinue the drug. Currently, there is no nomogram model for warfarin-associated bleeding risk. The aim of this study was to develop a risk-prediction nomogram model for warfarin-related major and clinically relevant non-major (CRNM) bleeding. A total of 280 heart disease outpatients taking warfarin were enrolled, 42 of whom experienced major or CRNM bleeding at the one-year follow-up. The Least Absolute Shrinkage and Selection Operator regression model was employed to identify potential predictors. Backward stepwise selection with the Akaike information criterion was used to establish the optimal predictive nomogram model. The receiver operating characteristic (ROC) curve, calibration plot, Hosmer–Lemeshow goodness-of-fit test, and decision curve analysis (DCA) were used to evaluate the performance of the nomogram. The nomogram consisted of four predictors: female (OR = 1.85; 95% CI: 0.91-3.94), TIA (OR = 6.47; 95% CI: 1.85-22.7), TTR (OR = 0.99; 95% CI: 0.97-1.00), and anemia (OR = 2.30; 95% CI: 1.06-4.84). The model had acceptable discrimination (area under the ROC curve = 0.68, 95% CI: 0.59-0.78), and was significantly better than the existing nine warfarin-related bleeding prediction scoring systems. The calibration plot and Hosmer–Lemeshow test (χ² = 7.557; P = .478) indicated well-calibrated data in the model. The DCA demonstrated good clinical utility. In this study, we developed a nomogram to predict the risk of warfarin-related major or CRNM bleeding. The model has good performance, allows rapid risk stratification of warfarin users, and provides a basis for personalized treatment.