Abstract Background To propose cut-off points for older adults’ weakness for upper and lower limbs muscle strength normalized by body size with the ratio standard/muscle quality and allometric scaling. Methods Ninety-four community-dwelling older adults (69.1% women) were assessed for 49 body-size variables (anthropometry, body composition and body indexes), handgrip strength (HGS), one maximum repetition measurement for knee extensors (1RM), isokinetic knee extension peak torque at 60°/s (PT), and six-minute walk test (6MWT). Ratio standard or muscle quality (muscle strength/body size) and allometric scaling (muscle strength/body size^b; when ^b is the allometric exponent) were applied for body-size variables significantly correlated with HGS, 1RM and PT. Cut-off points were computed according to sex based on mobility limitation (6MWT < 400 m) with ROC curve and Youden index. Results Absolute HGS, 1RM and PT cut-off points were not adequate because they were associated with body size (r > 0.30). But it was corrected with muscle strength normalization according to body size-variables: HGS (n = 1); 1RM (n = 24) and PT (n = 24). The best cut-off points, with the highest area under the curve (AUC), were found after normalization for men: HGS/forearm circumference (1.33 kg/cm, AUC = 0.74), 1RM/triceps skinfold (4.22 kg/mm, AUC = 0.81), and PT/body mass*height^0.43 (13.0 Nm/kg*m^0.43, AUC = 0.94); and for women: HGS/forearm circumference (1.04 kg/cm, AUC = 0.70), 1RM/body mass (0.54 kg/kg, AUC = 0.76); and PT/body mass^0.72 (3.14 Nm/kg^0.72; AUC = 0.82). Conclusions Upper and lower limbs muscle weakness cut-off points standardized according to body size were proposed for older adults of both sexes. Normalization removes the effect of extreme body size on muscle strength (both sexes) and improves the accuracy to identify weakness at population level (for women, but not in men), reducing the risk of false-negative/positive cases.