Background and Aims Hepatocellular carcinoma (HCC) is the most prevalent type of liver malignancy, accounting for 80% of all primary liver cancer cases. Partial hepatectomy is widely considered to be the treatment of choice for HCC. However, post-hepatectomy liver failure (PHLF) is the most serious complication and the leading cause of perioperative death. Therefore, an accurate assessment of the risk of PHLF is particularly critical. Patients with large hepatocellular carcinoma have larger tumors (tumor diameter ≥5 cm) and more resected liver tissue, and are more likely to develop PHLF. Previous studies have used various methods to assess the risk of PHLF, including liver function, Child-Pugh classification, model for end-stage liver disease, albumin-bilirubin (ALBI), and aspartate aminotransferase-to-platelet ratio index score. However, no model has been developed for data on hepatectomy for large HCC. Therefore, this study aims to analyze the risk factors of PHLF in HCC patients with large tumor and to construct a preoperative nomogram prediction model to guide and optimize clinical decision-making.Methods The clinical data of 927 patients with large liver cancer who underwent radical hepatectomy in the First Affiliated Hospital of Anhui Medical University (721 cases, training cohort) and the Second Affiliated Hospital of Anhui Medical University (206 cases, validation cohort) from January 2018 to June 2023 were retrospectively collected. The patients' baseline data, laboratory examination, imaging data, and surgical information were collected. Univariate analysis combined with multivariate analysis was used to screen out the independent risk factors for inducing PHLF, and binary Logistic regression was used to construct a prediction model for PHLF. ROC, calibration, and clinical decision curves verified the model's performance.Results There were no significant differences in all preoperative data between the training and validation cohorts (P>0.05). Grade B or C PHLF occurred in 192 of 927 patients (20.7%), including 8 patients with grade C PHLF. Univariate and multivariate Logistic regression analyses were used to determine the independent risk factors of PHLF, including tumor diameter, ALBI score, liver cirrhosis, vascular tumor thrombus, and intraoperative blood loss. These factors were included in the Logistic regression analysis, and a nomogram model was constructed to predict PHLF. The nomogram model was validated, and the C-index of the nomogram was 0.757. The ROC curve analysis of the prediction probability of the model showed that the AUC of the training set was 0.757 (95% CI=0.703-0.811), and the AUC of the validation set was 0.779 (95% CI=0.702-0.863). The validation showed that the model had good predictive ability.Conclusions Tumor diameter, ALBI score, liver cirrhosis, vascular tumor thrombus, and intraoperative blood loss are independent risk factors for PHLF. The nomogram prediction model constructed in this study can accurately assess the risk of preoperative PHLF, which is helpful for better clinical management, reducing the occurrence of PHLF, and improving the postoperative prognosis of patients.