The purpose of this study was to assess the test-retest reliability of the OMNI Scale for Resistance Exercise (OMNI-RES). Ten men (mean (SD) age 21.3 (1.4) years) participated in an initial session during which a one-repetition maximum (1-RM) for the leg extension exercise was assessed. Each subject then participated in two identical experimental sessions (separated by one week) in which one repetition of the leg extension exercise was performed at 40, 50, 60, 70, 80 and 90% of the 1-RM. Active muscle perceived exertion was assessed using the OMNI-RES scale immediately after the lift at each intensity. Subjects were unaware of the weight lifted in all sessions. The order of intensities was randomly assigned in both experimental sessions. A two-way (Session X Intensity) ANOVA indicated a significant (p < 0.01) Intensity main effect, but no significant interaction or Session main effect. Intraclass correlation coefficients were moderate at 40% through 80% of the 1-RM (0.66, 0.53, 0.43, 0.54, 0.54) and low at 90% of the 1-RM (0.29). These findings indicate that the stability of perceived exertion estimated using the OMNI-RES scale varies depending upon the relative exercise intensity. Keyword(s): exercise/fitness, research