The authors examined rapid response among 108 patients with binge eating disorder (BED) who were randomly assigned to 1 of 4 16-week treatments: fluoxetine, placebo, cognitive-behavioral therapy (CBT) plus fluoxetine, or CBT plus placebo. Rapid response, defined as 65% or greater reduction in binge eating by the 4th treatment week, was determined by receiver operating characteristic curves. Rapid response characterized 44% of participants and was unrelated to participants' demographic or baseline characteristics. Participants with rapid response were more likely to achieve binge-eating remission, had greater improvements in eating-disorder psychopathology, and had greater weight loss than participants without rapid response. Rapid response had different prognostic significance and distinct time courses for CBT versus pharmacotherapy-only treatments. Rapid response has utility for predicting outcomes and provides evidence for specificity of treatment effects with BED.