BACKGROUND Type 2 myocardial infarction (T2MI), caused by a mismatch in myocardial oxygen supply and demand, is a common complication in sepsis patients. However, its diagnosis remains challenging due to non-specific clinical presentations and the difficulty in distinguishing it from other causes of troponin elevation. This diagnostic ambiguity creates an urgent need for reliable diagnostic tools, especially since the management of T2MI fundamentally differs from that of type 1 MI. OBJECTIVE To develop, validate, and evaluate a diagnostic model for type 2 myocardial infarction in sepsis patients. METHODS Model development and internal validation were conducted using data from sepsis patients in the Medical Information Mart for Intensive Care IV database. External validation was performed on the intensive care unit patients of Jiading District Central Hospital Affiliated Shanghai University of Medicine & Health Sciences. Variables were selected using backward elimination. Then, they were incorporated into a logistic regression model to construct the diagnostic score. Discriminative ability and calibration were evaluated by the area under the receiver operating characteristic curve and calibration curves, respectively; prediction results were displayed using a nomogram and forest plots. RESULTS The derivation cohort included 2,519 patients and was divided into a training set (n=1,763) and a test set (n=756). External validation was performed on a separate cohort of 70 patients. Multivariate logistic regression identified the following independent factors associated with T2MI: diabetes mellitus, coronary heart disease, respiratory failure, old myocardial infarction, acute respiratory distress syndrome, septic shock, invasive mechanical ventilation, blood urea nitrogen, prothrombin time, international normalized ratio, temperature, and age. The final model achieved an AUC of 0.81 (95% CI: 0.79–0.83) in the training set, 0.79 (95% CI: 0.76–0.83) in the test set, and 0.78 (95% CI: 0.65–0.91) in the external validation cohort. The calibration curves showed excellent agreement between the predicted and observed probabilities across all datasets, indicating good model calibration. Sensitivity, specificity, positive predictive value, and negative predictive value were consistently high across all datasets. CONCLUSIONS The developed T2MI diagnostic scoring system for sepsis patients demonstrated strong discrimination and calibration, accurately predicting T2MI risk and proving useful for clinical diagnosis. CLINICALTRIAL none