Abstract: OBJECTIVE To identify and analyze the components of Cordyceps from different origins by UPLC-Q-TOF-MS. METHODS Metabolomic research of Cordyceps were performed using UPLC-Q-TOF-MS and multivariate statistical analysis technique. RESULTS Nucleosides were accounted for a large proportion of the identified components from Cordyceps. There was significant metabolites difference among Cordyceps from different origins, especially those from Hunan, Hubei and Lijiang. Cordyceps from Lijiang contained more deoxyadenosine than others and several flavonoids were detected in the Hunan sample. The Cordyceps from different origins were clustered into different groups respectively by principal component analysis(PCA). Samples from Tibet, Sichuang and Qinghai were clustered closely, while those from Liangshan, Hunan, Hubei and Lijiang were scattered around. CONCLUSION This study indicated that UPLC-Q-TOF-MS coupled with multivariate statistics could provide quality control of Cordyceps.