Although it is crucial to know the motivation underlying engaging in various activities, including physical ones, there is no Serbian version of a questionnaire for assessing motivation for exercise. The aim of this study is to evaluate the internal consistency reliability of the Serbian version of the EMI-2 questionnaire (Markland & Ingledew, 1997) in the fitness population. The research sample consisted of 1087 participants who exercise in fitness centers in the territory of the Banja Luka region. The integrated measuring instrument is made up of two parts: the first, self-created part of the questionnaire was used to describe the participants, and the second, standardised EMI-2 questionnaire was used to evaluate motivation for exercise. Descriptive statistics were used to describe the participants and the subscales. The Cronbach's alpha test was used to test the reliability of the EMI-2 scales, and Pearson correlation was used among the EMI-2 subscales. The results indicate a good reliability of the subscales in the range 0.606-0.850 and an overall reliability of 0.932. Pearson's correlation coefficient indicates a statistically significant correlation between the subscales except for Social Recognition with Positive Health and Nimbleness. The Serbian version of the EMI-2 has satisfactory internal consistency reliability and interrelatedness of the subscales.