OBJECTIVE: To explore the association between physical activity levels and metabolic dysfunction-associated fatty liver disease (MAFLD) and the modifying effects of different types of obesity. METHODS: A cross-sectional study was conducted on 19925 participants recruited from the Chengdu sub-cohort of the Southwest China Natural Population Cohort. The participants were recruited between 2018 and 2019. The association between physical activity and MAFLD prevalence was examined using the inverse probability weighting (IPW) method based on the generalized propensity score (GPS). The odds ratios (OR) and the 95% confidence interval (CI) for moderate and vigorous physical activity were calculated using the mild physical activity group as a reference. A restricted cubic spline function was used to model the exposure-response relationship between physical activity and MAFLD risk. The potential modifying effects of obesity types on the association between physical activity and MAFLD were evaluated in male and female populations. RESULTS: The prevalence of MAFLD was 17.30%. Compared to those engaging in mild physical activity, individuals participating in vigorous and moderate physical activities had a lower risk of MAFLD, with OR (95% CI) being 0.76 (0.67, 0.86) and 0.85 (0.76, 0.94), respectively. The exposure-response relationship showed a nonlinear association between physical activity and MAFLD risks ( CONCLUSION: Moderate physical activity can significantly reduce the risk of MAFLD, and the obesity types can modify this association. It is recommended that individuals engage in approximately 20-70 METs-h/d of physical activity. For females with central obesity, physical activity should not exceed 37.5 METs-h/d, while for females with peripheral obesity, it should not exceed 30 METs-h/d.