Purpose This study aims to investigate the impact of the compositions and concentrations of α -linolenic acid (ALA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which are significant n-3 polyunsaturated fatty acids (n-3 PUFAs), on the anti-inflammatory properties of lipopolysaccharide (LPS)-stimulated mouse hepatocytes. Design/methodology/approach A mouse hepatocyte inflammation model was established using LPS to evaluate the anti-inflammatory effects of a combination therapy with EPA, ALA and DHA. Interleukin-10 (IL-10), lipoxin A4 (LXA4) and tumor necrosis factor-α(TNF- α ) were measured to evaluate the anti-inflammatory ability of hepatocytes. The compositions and concentrations of EPA, ALA and DHA were optimized by response surface methodology to obtain the best anti-inflammatory effect. Findings The optimal inflammatory model was established with 10 µg/mL LPS treatment for 24 h. Maximal cell viability was found at EPA concentration of 150 µM, ALA concentration of 400 µM and DHA concentration of 600 µM. The optimal concentrations of n-3 PUFAs for anti-inflammatory activity in mouse hepatocytes were obtained through single-factor experiments and response surface analysis, with EPA 141.41 µM, ALA 405.45 µM and DHA 551.76 µM. Under optimized conditions, the experimental results for TNF- α , IL-10 and LXA4 levels were close to the theoretical prediction, indicating that the mathematical model was extremely accurate and versatile. Originality/value This study systematically explored the specific effects of different ratios and concentrations of ALA, EPA and DHA on the anti-inflammatory ability of mouse hepatocytes, offering an experimental foundation for the advancement of novel anti-inflammatory medicines and nutritional supplements.