Exercise is beneficial for mental health in general, but no review has systematically assessed its potential transdiagnostic nature, i.e. whether it is beneficial across specific disorders. We performed a systematic umbrella review of meta-analyses of randomized controlled trials (RCTs) of exercise in participants with mental disorders defined according to the International Classification of Diseases (ICD) or the Diagnostic and Statistical Manual of Mental Disorders (DSM), assessing exercise's transdiagnostic nature with TRANSD criteria, including eight meta- analyses (six included in the TRANSD meta-analysis), encompassing 99 RCTs (n = 5,656) across 11 disorders. Moderate/vigorous aerobic exercise was an effective transdiagnostic intervention for disease-specific primary symptoms across 11 disorders (recurrent depressive disorder, social phobia, panic disorder, generalized anxiety disorder, post-traumatic stress disorder, brief psychotic disorder, schizophrenia, schizoaffective disorder, delusional disorder, schizophreniform disorder, attention-deficit/hyperactivity disorder) and four spectra (depressive disorders, anxiety disorders, schizophrenia-spectrum disorders, neurodevelopmental disorders) with a medium effect size (SMD =-0.67, 95%CI =-0.84,-0.50). Moderate/vigorous aerobic exercise also improved cognition across two disorders (schizophrenia, attention-deficit/hyperactivity disorder) and two spectra (schizophrenia- spectrum disorders, neurodevelopmental disorders), with a large effect size (SMD = 0.92, 95%CI = 0.52, 1.33). According to TRANSD criteria, moderate/vigorous aerobic exercise is a transdiagnostic intervention to improve disease-specific primary symptoms of 11 mental disorders, and cognition in two mental disorders.