Sarcopenia is known to be associated with non-alcoholic fatty liver disease (NAFLD). However, few studies have revealed the association between muscle strength and prevalence of NAFLD. We investigated the association by using relative handgrip strength in a nationwide cross-sectional survey. The participants were recruited from the Korean National Health and Nutrition Examination Surveys (KNHANES). A total of 27,531 subjects from the KNHANES were selected in our study. We used normalized handgrip strength, which is called relative handgrip strength. The index was defined as handgrip strength divided by BMI. These subjects were divided into quartile groups according to relative handgrip strength. NAFLD was defined as hepatic steatosis index >36. Multinomial logistic regression was analysed to investigate the association between relative handgrip strength with prevalence of NAFLD. The mean age of study population was 45.8 ± 0.3 in men, and 48.3 ± 0.2 in women. The proportion of males was 37.5%. In multiple linear regression, relative handgrip strength was inversely associated with HSI index (Standardized β = −0.70; standard error (SE), 0.08; p < 0.001 in men, Standardized β = −0.94; standard error (SE), 0.07; p < 0.001 in women). According to the logistic regression model, the prevalence of NAFLD decreased with quartile 4 groups in relative handgrip strength, compared with quartile 1 groups (OR 0.42 [0.32–0.55] in men; OR 0.30 [0.22–0.40] in women). Relative handgrip strength, used as a biomarker of sarcopenia, is independently inversely associated with NAFLD.