Children and adolescents with neurodevelopmental disorders experience mental health issues and are encouraged to be brought together in real-life treatment. However, physical activity, which is a promising method for boosting mental health, has only been examined in children and adolescents with certain types of neurodevelopmental disorders. To examine the association of physical activity interventions with mental health (ie, cognitive function, psychological well-being, internalizing, and externalizing problems) in children and adolescents with neurodevelopmental disorders and to identify possible moderators. Studies were searched from inception to May 2023 through Web of Science, PsycINFO, SPORTDiscus, MEDLINE, CINAHL, and ERIC. Randomized clinical trials or nonrandomized designs applying physical activity interventions and reporting at least 1 mental health outcome in children and adolescents aged 5 to 17 years with neurodevelopmental disorders were included. Two independent reviewers selected and assessed the studies. Random multilevel meta-analysis using Hedges g was performed. Data extraction and risk-of-bias assessment were conducted by multiple reviewers. Heterogeneity, publication bias, sensitivity analysis, and moderator analysis were examined to substantiate the results. The main outcomes were mental health related to cognitive function, psychological well-being, internalizing, or externalizing problems measured by neurocognitive tasks or subjective questionnaires. A total of 76 studies involving 3007 participants were included in systematic review, 59 of which were used for meta-analysis. The findings indicated that physical activity interventions were associated with significant benefits for overall mental health (g, 0.67; 95% CI, 0.50-0.85), cognitive function (g, 0.74; 95% CI, 0.53-0.95), psychological well-being (g, 0.56; 95% CI, 0.16-0.96), internalizing (g, 0.72; 95% CI, 0.34-1.10), and externalizing problems (g, 0.58; 95% CI, 0.28-0.89). Moderators were also identified: frequency, total sessions, and total duration for overall mental health; total sessions, and total duration for cognitive function; session duration and frequency for psychological well-being; physical activity type for internalizing problems; and session duration for externalizing problems. Moderator analyses showed that type of neurodevelopmental disorder did not modify the associations between physical activity and overall mental health or its subgroups. The findings in this study suggest that children and adolescents with different types of neurodevelopmental disorders may be grouped together when performing physical activity interventions, which were confirmed to be beneficial to overall mental health and its subgroups in this new diagnostic population, but that physical activity interventions should be tailored when targeting different mental health domains.