Objective: To explore the application and predictive value of FGF8 as a biomarker in sepsis. Methods: A total of 44 patients with sepsis who were diagnosed and treated in our hospital from December 2024 to February 2026 were selected as the study subjects, and another 44 volunteers who underwent health check-ups in our hospital during the same period were selected as the control group. Clinical data and levels of FGF8, procalcitonin (PCT), C-reactive protein (CRP), white blood cell count (WBC), red blood cell count (RBC), hemoglobin (Hb), and platelet (PLT) in peripheral venous blood were collected from both groups. Univariate analysis was performed, and variables with statistical significance in the univariate analysis were included in the multivariate logistic regression analysis. ROC curves were plotted to explore the predictive value of FGF8 as a biomarker in sepsis. Results: Univariate analysis revealed that APACHE II score, SOFA score, FGF8 level, and PLT were statistically significant. After inclusion in the multivariate logistic regression analysis, APACHE II score, SOFA score, and FGF8 level were positively correlated with sepsis, while PLT was negatively correlated with sepsis. Subsequent ROC curve plotting showed that FGF8 had a sensitivity of 82.67%, a specificity of 78.50%, and an AUC of 0.840 in sepsis, indicating certain predictive value. Conclusion: FGF8, as a biomarker, has certain predictive value in sepsis, providing a potential new candidate marker for the auxiliary diagnosis of sepsis, which deserves clinical attention.