Background: Post-hepatectomy liver failure (PHLF) is a leading cause of mortality after major liver resection. Accurate preoperative risk assessment is essential, yet current methods have limitations. Gadoxetic acid-enhanced MRI (Gd-EOB MRI) enables both morphological and functional evaluation of the liver. The aim of this study was to evaluate the efficacy of the hepatic uptake index (HUI) obtained from routine preoperative Gd-EOB MRI for identifying patients at risk of severe PHLF. Methods: This observational retrospective multicentre study included 292 patients who underwent major hepatectomy between 2010 and 2020 in Sweden, Denmark, and Finland. Preoperative Gd-EOB MRI was performed for each patient and the HUI, hepatic uptake index of the standardized future liver remnant (sFLR-HUI), and Model for End-Stage Liver Disease Version 3 (MELD 3) score were evaluated. Statistical analyses included logistic regression and receiver operating characteristic (ROC) curve assessment to determine cut-off values and discriminative accuracies for severe PHLF (International Study Group of Liver Surgery grades B and C). Results: Among the 292 patients, 25 (8.6%) developed severe PHLF. Patients with severe PHLF had significantly lower HUI and sFLR-HUI values (P < 0.001). The HUI demonstrated superior discriminative performance for severe PHLF (area under the curve (AUC) 0.758) compared with volume-only assessments, such as the standardized future liver remnant (sFLR) (AUC 0.628). Combining the HUI with the MELD 3 score improved performance further (AUC 0.803). Conclusion: The HUI obtained from routine Gd-EOB MRI outperforms volume-based biomarkers (sFLR) for identification of patients at risk of severe PHLF. Incorporating image-derived functional assessments, such as the HUI, with independent biomarkers, such as the MELD 3 score, may optimize preoperative risk stratification for severe PHLF and improve outcomes after major hepatectomy.