Self-efficacy is a major motivational factor that explains the effectiveness of people's functioning alongside cognitive and personality characteristics. However, our knowledge regarding self-efficacy among gifted students is underdeveloped. Therefore, in this meta-analysis, we examined the difference in self-efficacy between gifted and non-gifted students. In total, 25 studies (m = 70; N = 42,736) were analyzed using a three-level meta-analytical approach. The unconditional model revealed the overall mean effect size was g = 0.54, SE = 0.107, 95 % CI 0.326, 0.747, p < .001, which indicates that gifted students had moderately higher self-efficacy than their non-gifted peers. However, we observed symptoms of publication bias and small-study effects. After correcting for these effects, the mean effect size was estimated as g = 0.26, SE = 0.109, 95 % CI 0.047, 0.476, p = .032, thus indicating a small to moderate difference between the groups. Moderator analyses showed that the self-efficacy domains (academic and socio-emotional), sex, and age moderated the overall effect size obtained, while culture (East vs. West) and the general, mathematical, and verbal self-efficacy domains did not significantly differentiate the obtained effects. These findings support an argument against exaggerations of differences between gifted students and their non-gifted peers. •A meta-analysis clarifying self-efficacy in gifted versus nongifted students•Analyzes variations in effect size by culture, sex, age, and self-efficacy domain•Gifted students had moderately higher self-efficacy than nongifted students.•Sex, age, and self-efficacy domains were significant moderators in the current study.