OBJECTIVE To provide reference for Smilax glabra quality control. METHODS UPLC method was established to simultaneously determine the contents of 5-O-caffeoylshikimic acid， neoastilbin， astilbin， neoisoastilbin， isoastilbin， engelitin， resveratrol and isoengelitin in 20 batches of S. glabra from different areas （No. S1-S20）. The quality evaluation of 20 batches of samples was performed by chemometrics； the differential biomarkers that affected the quality of S. glabra were screened. RESULTS The measured 8 components had good linear relationship within the range of measured concentration （r≥0.999 6）. RSDs of precision， repeatability and stability tests （24 h） were all lower than 2.00% （n＝6）. The average recoveries varied between 97.60% and 106.40% （RSDs were all lower than 2.00%， n＝6）. Cluster analysis showed that the samples produced in Zhejiang （S1-S5） and Jiangxi （S6-S10） were clustered into one category； the samples produced in Hunan （S11-S15） were clustered into one category； the samples produced in Yunnan （S16-S20） were clustered into one category. Principal component analysis showed that the first two principal components could represent 85.60% information of 8 components in S. glabra. Partial least squares-discriminant analysis showed that variable importance projection values of 5-O-caffeoylshikimic acid， astilbin， isoastilbin and neoastilbin were all greater than 1. CONCLUSIONS There are differences in the contents of the above 8 components in S. glabra from different origins； 5-O-caffeoylshikimic acid， astilbin， isoastilbin and neoastilbin may be differential markers affecting the quality of S. glabra.