Objective To construct a 28d mortality risk prediction model for sepsis patients based on the Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE II) score, validate its predictive performance, and provide an auxiliary tool for clinical precision assessment and risk stratification. Methods A retrospective study was conducted on 186 patients diagnosed with sepsis in Intensive Care Unit, Shangqiu First People’s Hospital from January 2023 to June 2025. APACHE Ⅱ score related indicators, lactic acid levels, underlying diseases, mechanical ventilation, and other clinical data were collected, and 28d survival outcomes were recorded. Multivariate Logistic regression analysis was used to screen for independent risk factors, and a nomogram model was constructed. The discriminant power, calibration degree, and clinical utility of the model were evaluated through C-index, calibration curve, and decision curve analysis (DCA), and internal Bootstrap validation was conducted. Temporal validation was conducted by dividing patients into a derivation cohort (from January 2023 to December 2024, n=138) and a temporal validation cohort (from January 2025 to June 2025, n=48). Results Multivariate Logistic regression analysis showed that APACHE II score, lactic acid level, mechanical ventilation and diabetes/heart failure were independent factors affecting 28d mortality in patients with sepsis (P<0.05). The constructed nomogram model had a C-index of 0.872, and the calibration curve fits well. DCA showed a high clinical net benefit. The internal validation C-index was 0.860, indicating good model stability. In the temporal validation cohort, the model maintained good discrimination with the C-index of 0.852. Conclusion The nomogram model based on APACHE Ⅱ score can accurately predict the 28d mortality risk of sepsis patients, with good discriminative ability, calibration consistency, and clinical practicality, and is suitable for early clinical risk assessment and intervention decision-making.