Background: Exercise is recommended to protect physical health among people with severe mental illness and holds the potential to facilitate long-term recovery. An inclusive exercise community provides an opportunity for life skill training and social connectedness and may reduce the experience of loneliness and internalized stigmatization which together may improve personal recovery. Using a pragmatic randomized design, we aim to examine the effectiveness of a gym-based exercise intervention tailored to young adults in antipsychotic treatment (i.e., Vega Exercise Community) compared to usual care. It is hypothesized that the Vega Exercise Community will be superior to usual care for personal recovery at four months. Methods: The trial will be conducted at four sites in Denmark from which 400 participants, aged 18 to 35 years, who are in current treatment with antipsychotic medications for the management of schizophrenia spectrum or affective disorders, will be recruited. Participants will be randomized (2:1) to Vega Exercise Community or usual care. Vega Exercise Community includes three weekly group-based exercise sessions hosted in commercial functional training centers delivered by certified Vega instructors. After four months, participants in Vega Exercise Community will be randomized (1:1) to minimal versus extended support with regards to sustained physical activity. Data will be collected at baseline, four, six and 12 months. The primary outcome is personal recovery assessed by Questionnaire about the Process of Recovery at four months. Behavioral symptoms, health-related quality of life, metabolic health, and program costs will be evaluated to further determine the effectiveness and cost-effectiveness of the Vega Exercise Community. Finally, the quality of life and physical and mental health of the participants’ primary relative will be evaluated. Discussion: The results of this trial may have important implications for health, sustained physical activity, and recovery for individuals in treatment with antipsychotics. Given the pragmatic design, positive results may readily be implemented by mental health care professionals to promote exercise as an integrated part of treatment of severe mental illness. Trial registration: Clinical Trials.gov (NCT05461885, initial registration June 29th, 2022). WHO Universal Trial Number (UTN): U1111-1271–9928.