Abstract Background: Despite advancements in antiretroviral treatment (ART), a significant number of HIV patients continue to die due to the infection. In developing countries like Ethiopia, there is no prediction model to assess mortality risk in patients receiving ART. This study was designed to Develop a mortality risk prediction model in patients receiving ART at Felege Hiwot. Comprehensive Specialized Hospital (FHCSH). Methods: A retrospective-cohort study was conducted at FHCSH to develop a risk prediction. model for mortality in patients receiving ART. The sample size was calculated using the rule of thumb and included 777 participants. Secondary data were utilized, and data entry and analysis were performed using Epi-Info and R statistical software. Logistic regression analysis was conducted to identify the prognostic determinants of mortality. Model performance was assessed. using the receiver operating characteristic (ROC) curve (AUC), along with calibration plots. The Mortality risk prediction was presented in the form of a nomogram. Model accuracy and risk Estimates were assessed via internal validation using the bootstrapping method, and the clinical and The public health benefit of the model was evaluated by decision curve analysis (DCA). Results: The incidence of HIV mortality was 12.6% (95% CI: 10.2%-15.1%). Six prognostic determinants: WHO clinical staging, functional status, baseline regimen, duration of treatment, Adherence and comorbidities were used to develop the mortality risk prediction nomogram. The AUC of the original model was 0.968 (95% CI: 0.951-0.983), while the AUC of the reduced model demonstrated excellent discriminatory power at 0.967 (95% CI: 0.947-0.987), with calibration fitting well. At a cut-off value of 0.1667 (Youden index), the model achieved an accuracy of 93.3%, a sensitivity of 87.8%, and a specificity of 94.1%. The DCA highlighted its clinical and public health importance within the range of 0.1 to 0.8 thresholds. Conclusions: In this study the incidence of mortality is somewhat high as compared with the previous studies, and six prognostic determinants were used to develop the nomogram to predict risk of mortality. The AUC curve showed an excellent discrimination, with strong sensitivity and specificity, helping healthcare providers to forecast mortality risk in patients receiving ART. However, before implementing in a clinical setting, a multicenter, large-scale, prospective cohort Studies are warranted to validate its generalizability and effectiveness.