OBJECTIVE: Total glycoside of peony (TGP) enhances methotrexate efficacy and attenuates its hepatotoxicity in rheumatoid arthritis, but the mechanisms remain unclear. This study investigates the mechanisms of TGP against methotrexate-induced liver injury through a network pharmacology-based approach. METHODS: A liver injury model was established in CD-1 mice by intraperitoneal injection of 20 mg/kg methotrexate. TGP and the positive control drug silybin were used to intervene in the methotrexate-induced liver injury model in mice. Serum ALT and AST activities, liver index test and histopathology was detected to evaluate the effects of the treatment on methotrexate-induced liver injury. Additionally, network pharmacology and serum metabolomics were employed to predict the mechanisms of TGP in treating methotrexate-induced liver injury. Experimental validation was conducted by RT-PCR, ELISA and Western blot. RESULTS: TGP effectively alleviated the liver index and pathological liver damage induced by methotrexate and reduced the liver injury markers, serum ALT and AST, showing effects comparable to those of the positive control drug silybin. Network pharmacology predicted that the key targets and key signaling pathways of TGP in treating methotrexate-induced liver injury are closely associated with inflammatory response. Furthermore, serum metabolomics and network pharmacology analysis indicated a close association between effects of TGP on methotrexate-induced liver injury and arachidonic acid pathway. Experimental validation results confirmed that the expression levels of IL-6, TNF and COX-2 in liver tissues were significantly elevated, with the activation of the PI3K/AKT, MAPK, and NFκB pathways. TGP intervention can reverse these changes to a certain extent. CONCLUSION: TGP treatment effectively mitigates methotrexate-induced liver injury, and its mechanism is closely associated with the inhibition of hepatic inflammatory responses.