GC-MS and FTIR analysis, combined with multivariate analysis methods such as PCA and PLS, were used to evaluate and identify moldy tobacco leaves. GC-MS analysis screened 9 markers of tobacco leaf mold, including 2-ethylhexanol. The linear discriminant equation constructed by 9 key compounds, such as 4-hydroxybutyrolactone, can accurately identify tobacco leaf mold, with an initial validation accuracy of 100% and a cross-validation accuracy of 98.7%. FTIR studies have shown that the moldy process of tobacco leaves consumes a large amount of carbohydrates, proteins, and lipids. GC-MS and FTIR, combined with PCA and PLS-DA respectively, can effectively distinguish moldy tobacco leaves.