Objective To assess the differences in behavior and molecular mechanism of C57BL/6 mice subjected to repeated corticosterone injection (CORT) or chronic unpredictable mild stress (CUMS), and to provide a theoretical reference for antidepressants screening and evaluation. Methods Thirty male C57BL/6 mice were divided into control group, CORT group and CUMS group. During the 3 week stress period, body weights of mice were measured every 3 days. After stress exposure, the open-field test, force swimming test and tail-suspension test were used to evaluate the behavioral changes, with serum corticosterone measured by ELISA. Histological studies were carried out the hippocampal neuron damage with Nissl staining, while the expressions of brain CRH, BDNF, p-CREB and p-ERK protein or gene transcripts were analyzed by Western blotting or PCR. Results Compared with the control group, the number of grooming was significantly decreased in the CORT group, with no significant changes in frequency of crossing and rearing. In the CUMS group, the numbers of rearing and crossing were significantly decreased, while the frequency of grooming was not changed. In the force swimming and tail suspension tests, the time of immobility was significantly increased in both CORT and CUMS groups compared with the control group. Serum corticosterone levels were significantly higher in CORT and CUMS groups than control group. Comparing between the two model and the control groups, there was no significant difference in the thymus index, while the spleen index in the CORT group was significantly decreased. The density of CA1, CA3 and dentate gyrus regions Nissl stained neurons reduced in both CUMS and CORT group, especially in CORT group. Through PCR detection, levels of brain CRH mRNA in both CORT and CUMS group were significantly higher than the control group. Levels of BDNF, p-CREB and p-ERK protein were decreased in the CORT and CUMS groups relative to control, whereas CRH protein levels were higher in the former two groups. Conclusion Both the CORT and CUMS models present depression behaviors, which appears to reflect dysfunction of the hypothalamic-pituitary-adrenal (HPA) axis. There is no significant difference between CORT and CUMS models in behavior alteration, hippocampal formation and protein expression of BDNF-p-CREB and ERK signaling pathway. In conclusion, the CORT model could be a useful model of depression and might be applied for mechanism research and antidepressant screening. The CORT model has an advantage of simple operation and shorter modeling cycle over the CUMS model.