This is the first trial to investigate the outcome of tailored and ACT-influenced, cognitive behavioral Internet treatment for eating disorder psychopathology, and the relation between knowledge acquisition and outcome. This study utilized a randomized controlled design, with computer-based allocation to treatment or waiting list control group. Participants were recruited via advertisements in social media and newspapers in Sweden. Participants fulfilling the criteria for bulimia nervosa (BN), orEating Disorder Not Otherwise Specified (EDNOS), with a BMI above 17.5, were enrolled in the study (N = 92). The treatment group received an Internet-based, ACT-influenced CBT intervention, developed by the authors, for eating disorders. The treatment lasted 8 weeks, and was adapted to the participants individual needs. A clinician provided support. The main outcome measures were eating disorder symptoms and body shape dissatisfaction. Intent-to-treat analysis showed that the treatment group (n = 46) improved significantly on eating disorder symptoms and body dissatisfaction, compared with the waiting list control group (n = 46), with small to moderate effect sizes (between group effects, d = 0.35-0.64). More than a third of the participants in the treatment group (36.6%), compared to 7.1% in the waiting list control condition, made clinically significant improvements. Results showed a significant increase in knowledge in the treatment group compared to the waiting list control group (between group effect, d = 1.12), but we found no significant correlations between knowledge acquisition and outcome (r = -0.27 to -r = 0.23). The results provide preliminary support for Internet-based, tailored, and ACT-influenced treatment, based on CBT for participants with eating disorder psychopathology.