Weight gain is a major adverse effect of several second-generation antipsychotic medications. Rimonabant is a cannabinoid-1 receptor antagonist that promotes weight loss in the general population. We conducted a 16-week, double-blind, placebo-controlled study of rimonabant (20 mg/d) in people with schizophrenia or schizoaffective disorder, based on the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria, who were clinically stable on second-generation antipsychotics. Participants had a body mass index of 27 kg/m or higher with hyperlipidemia or body mass index of 30 kg/m or higher, and no current substance abuse/dependence (except nicotine), more than weekly cannabis use, or recent depressive symptoms/suicidality. An exercise and dietary counseling group was offered weekly. Target enrollment was 60; the trial was terminated early because of withdrawal of rimonabant from the European market. Fifteen participants were randomized (7 rimonabant, 8 placebo); 5 completed in each group. Rimonabant was associated with a greater reduction in Brief Psychiatric Rating Scale total score versus placebo (mean ± SE difference, -1.9 ± 0.8, P = 0.02), driven by differences in the Brief Psychiatric Rating Scale anxiety/depression (-1.4 ± 0.35, P = 0.0004) and hostility (-0.7 ± 0.3, P = 0.02) factors. Group differences were not significant for the Calgary Depression Scale total score (P = 0.24), Scale for the Assessment of Negative Symptoms total score (P = 0.13), weight, blood pressure, or fasting lipids or glucose. Rimonabant was well tolerated with no significant adverse events. No significant weight loss, metabolic effects, or adverse psychiatric effects were associated with the cannabinoid-1 receptor antagonist rimonabant in this small sample of people with schizophrenia. The endocannabinoid system remains a promising target for pharmacotherapy of schizophrenia and obesity.