Objective To establish a prediction model to assess the risk of death in patients with Klebsiella pneumoniae (Kp) infection, and to provide a reference for the development of targeted prevention and control strategies for nosocomial infections. Methods Patients with Klebsiella pneumoniae(Kp) infection at the Affiliated Hospital of Guizhou Medical University from January 2020 to December 2022 were used as the training dataset. LASSO-logistic regression was applied to identify independent risk factors for death in Kp-infected patients, and a mortality risk nomogram was constructed. Patients with Kp infection from January 2023 to December 2023 were used as the validation dataset. Model discrimination, accuracy, and clinical utility were comprehensively evaluated using the area under the receiver operating characteristic curve (ROC), the calibration curve, and the decision analysis curve. Results A total of 1 972 Kp-infected patients were included in the training dataset, of which 234 Kp-infected patients died with a mortality rate of 11.87%. A total of 1 148 Kp-infected patients were included in the validation dataset, of which 126 Kp-infected patients died with a mortality rate of 10.98%. The results of LASSO-logistic regression modeling showed that patients aged >70 years, carbapenem resistance, mechanical ventilation, indwelling urinary catheter, diabetes mellitus, hematologic infections, white blood cell count >10×10⁹/L and hemoglobin <115 g/L were all associated factors for death in patients with Kp infection (OR=24.6, 95% CI: 16.56-39.38; OR=2.44, 95% CI: 1.70-3.52; OR=4.97, 95% CI: 3.08-8.21; OR=2.99, 95% CI: 1.75-5.19; OR=18.24, 95% CI: 9.27-38.12; OR=2.75, 95% CI: 1.69-4.49; OR=4.18, 95% CI: 2.88-6.09; OR=2.65, 95% CI: 1.79-3.93). The areas under the ROC curve of the training dataset and validation dataset were 0.900 (95% CI: 0.876-0.924) and 0.843 (95% CI: 0.812-0.874), respectively. The calibration curve was close to the baseline curve, and the decision analysis curve was close to the upper right corner. Conclusion By analyzing the factors related to the death of Kp-infected patients, we constructed a nomogram prediction model of the risk of death of Kp-infected patients, which has good accuracy, discrimination, and clinical utility, and according to which targeted nosocomial infection prevention and control measures can be implemented.