Background: Alcoholic hepatitis is a severe form of liver disease associated with high mortality rates. Early identification of high-risk patients is crucial for optimizing treatment strategies and improving outcomes.  Aims: This study focused to evaluate the accuracy of the Glasgow Alcoholic Hepatitis Score (GAHS) and the Modified Discriminant Function (mDF) score in predicting mortality among patients with alcoholic hepatitis and to compare their predictive performance.  Methods: A longitudinal cross-sectional study was conducted among 140 patients with alcoholic hepatitis. Demographic, clinical, and laboratory data were collected, and GAHS and mDF scores were calculated. Univariate along with multivariate logistic regression analyses were carried out to identify predictors of mortality.  Results: The majority of participants were males (92.9%), with mean age of 42.5 ± 9.8 years. The GAHS demonstrated superior performance compared to mDF, with a sensitivity of 92.3%, specificity of 96.5%, PPV of 86.2%, NPV of 98.2%, and AUC of 0.95 (p=0.02). Univariate along with multivariate logistic regression analyses confirmed that GAHS, mDF, age, and serum bilirubin are significant mortality predictors in alcoholic hepatitis patients. Mortality rates increased with the severity of liver disease (mild: 5%, moderate: 15%, severe: 30%, p < 0.001). Conclusion: GAHS and mDF scores are effective tools for predicting death in alcoholic hepatitis patients, with GAHS showing better predictive ability. Early identification of highrisk patients using these scores and implementing appropriate therapeutic interventions can improve outcomes in this patient population.