OBJECTIVE: This study aimed to develop and validate a nomogram to predict the risk of sepsis in non-traumatic subarachnoid hemorrhage (SAH) patients using data from the MIMIC-IV database. METHODS: A total of 803 SAH patients meeting the inclusion criteria were randomly divided into a training set (563 cases) and a validation set (240 cases). Independent prognostic factors were identified through forward stepwise logistic regression, and a nomogram was created based on these factors. The discriminative ability of the nomogram was assessed using the area under the receiver operating characteristic curve (AUC) and compared with the SOFA score. The model's consistency was evaluated using the C-index, and the improvement in performance over the SOFA score was calculated using integrated discrimination improvement (IDI) and net reclassification improvement (NRI). RESULTS: Five independent predictive factors were identified through LASSO regression analysis: mechanical ventilation, hyperlipidemia, temperature, white blood cell count, and red blood cell count. The AUC of the nomogram in the training and validation sets were 0.854 and 0.824, respectively, both higher than the SOFA score. NRI and IDI results indicated that the nomogram outperformed the SOFA score in identifying sepsis risk. Calibration curves and the Hosmer-Lemeshow test demonstrated good calibration of the nomogram. Decision curve analysis showed that the nomogram had higher net benefit in clinical application. CONCLUSION: The nomogram developed in this study performed excellently in predicting the risk of sepsis in SAH patients, surpassing the traditional SOFA scoring system, and has significant clinical application value.