Abstract: Based on the establishment of high performance liquid chromatography(HPLC) fingerprint for cassava leaf flavonoids, similarity evaluation(SE), cluster analysis(CA) and principal component analysis (PCA) were used to evaluate the quality of cassava leaf flavonoids and determine the key components that led to their quality differences. This study indicated that the results of method validation of HPLC fingerprint met technical standards. Five common peaks was verified, and the similarities of 18 leaf flavonoids samples from six cassava varieties at three planting were higher than 0.88, simultaneously the sample score points were located within the tolerance ellipse of PCA, which showed that the quality of cassava leaf flavonoids from different cultivars and planting ages was relatively consistent. CA showed that samples from different cultivars and different planting ages could also cluster together, indicating that the quality of cassava leaf flavonoids was influenced by both cultivars and planting ages. Three principal components with the characteristic root cumulative contribution rate reaching 86.748% were screened out by PCA results. The composite score of SC10-180 d, SC205-120 d, SC205-180 d were the highest (2.2 above), and they had the best quality. The peaks of rutin X3, hesperidin X6 and amentoflavone X15 were important reasons for the quality difference of samples. The analysis method and result of fingerprint combined with SE, CA, PCA in this paper can provide a new approach for the quality assessment of cassava leaf flavonoids and also be good for their quality control.