Objective To compare two method of Mycobacterium tuberculosis culture (the BACTEC MGIT 960rapid culture system and the L?wenstein-Jensen plates (L-J plates) culture method) and to explore the value of theBACTEC MGIT 960 rapid culture system in a Mycobacterium tuberculosis (M.tb)-infected mouse model. Methods Theexperimental groups were divided into BACTEC MGIT 960 and L-J culture groups. Thirty-six female C57BL/6 mice wereinfected with 1. 0×106 colony forming units (CFU) / mL H37Rv via the tail vein. Eighteen mice were treated with rifampicinfor 1 week after infection. Eighteen mice were injected with the same amount of phosphate buffer (PBS) as a control. Thelung, spleen, and liver homogenates of 36 mice were cultured by BACTEC MGIT 960 rapid culture and L-J culture.Results In the BACTEC MGIT 960 culture system group, the time for a positive result in the liver, lung, and spleentissue cultures in the RIF treatment group was (187. 11±10. 20) h, (347. 22±12. 70) h, and (276. 39±13. 09) h,respectively, and in the control group it was (142. 50± 11. 70) h, (251. 67± 16. 63) h, and (230. 28 ± 7. 22) h,respectively. There were significant differences between the two groups ( P < 0. 001). In the L-J plates group, the bacterialload of the liver, lung, and spleen tissue cultures in the RIF treatment group was (5. 15±0. 15) log₁₀ CFU, and (3. 30±0. 23) log₁₀ CFU, and (3. 40±0. 25) log₁₀ CFU, and in the control group it was (5. 90±0. 25) log₁₀ CFU, (3. 88±0. 31)log₁₀ CFU, and (4. 15 ± 0. 30) log₁₀ CFU. There were significant differences between the two groups ( P < 0. 001).Conclusions The culture results are consistent between the BACTEC MGIT 960 and the L-J plates culture methods.However, the time to detect a positive result is reduced by half on average, and the discrimination is higher when using the BACTEC MGIT 960 culture method in an animal model.