Hepatocellular carcinoma (HCC) is a major global health concern, being the sixth most common cancer and the third leading cause of cancer deaths worldwide. Less than 30% of HCC patients are eligible for curative treatment, primarily due to diagnosis at advanced stages. This emphasizes the importance of early detection in improving survival outcomes. In this study, we investigated the methylation levels of certain genes and miRNAs in liquid biopsy and developed a methyl predictive model (MPM-8G). The AUC for MPM-8G was found to be significantly higher than that for AFP (alpha-fetoprotein) alone. When MPM-8G and AFP were combined, the AUC increased notably, indicating that the combined use of MPM-8G and AFP offers superior diagnostic performance and enhances the accuracy of HCC detection. Furthermore, the combination of MPM-8G and AFP proved to be a powerful tool for early diagnosis of HCC. This study successfully identified differences in the methylation levels of certain genes and miRNAs in liquid biopsy from HCC patients, leading to the construction of a predictive model for early diagnosis. The impressive performance of these methylation markers underscores their potential for further clinical application in the management of HCC.