Background and Aim: A prediction model of hepatocellular carcinoma (HCC) risk in patients with chronic liver diseases, based on age, male sex, albumin-bilirubin, and platelets (aMAP), has been previously reported. We validated the aMAP score and compared its performance to those of other risk scores in an independent at-risk cohort.Methods: Treatment-naïve patients with chronic hepatitis B-related compensated cirrhosis who received entecavir or tenofovir monotherapy for at least 12 months were enrolled in this study. The performances of the aMAP and other HCC risk scores were assessed using Harrell's c-index, and predefined cut-off values were evaluated using survival analysis.Results: Of the 1,042 patients, 131 (12.6%) developed HCC during a median follow-up of 41 months. The aMAP score provided the highest Harrell's c-index (0.724), followed by CAMD (0.719), mPAGE-B (0.719), and PAGE-B (0.695). The 5-year cumulative HCC probabilities were 2.9% for patients with a low aMAP score (<50), 11.2% for patients with an intermediate aMAP score (50–60), and 27.9% for patients with a high aMAP score (>60). Using both aMAP and mPAGE-B, 11.6% of patients were identified as low risk with a negative predictive value of 98.2% for not developing HCC within 5 years. Patients with aMAP >60 and diabetes exhibited an extremely high risk of HCC, with a cumulative incidence of 49.3% at 5 years. The predictive performance of aMAP with a reassessment at 1 year after the initiation of antiviral therapy outperformed the predictive performance of aMAP at enrollment.Conclusions: The aMAP score accurately predicted the risk of HCC in at-risk patients with compensated cirrhosis undergoing antiviral therapy. A combination of the aMAP score and diabetes status could further stratify the risk of HCC.