Objective: Our aim in this randomized controlled trial was to investigate the effects on global tinnitus severity of 2 Internet-delivered psychological treatments, acceptance and commitment therapy (ACT) and cognitive behavior therapy (CBT), in guided self-help format. Method: Ninety-nine participants (mean age = 48.5 years; 43% female) who were significantly distressed by tinnitus were recruited from the community. Participants were randomly assigned to CBT (n = 32), ACT (n = 35), or a control condition (monitored Internet discussion forum; n = 32), and they were assessed with standardized self-report measures (Tinnitus Handicap Inventory; Hospital Anxiety and Depression Scale; Quality of Life Inventory; Perceived Stress Scale; Tinnitus Acceptance Questionnaire) at pre-, posttreatment (8 weeks), and 1-year follow-up. Results: Mixed-effects linear regression analysis of all randomized participants showed significant effects on the primary outcome (Tinnitus Handicap Inventory) for CBT and for ACT compared with control at posttreatment (95% CI [-17.03, -2.94], d = 0.70, and 95% CI [-16.29, -2.53], d = 0.68, respectively). Within-group effects were substantial from pretreatment through 1-year-follow-up for both treatments (95% CI [-44.65, -20.45], d = 1.34), with no significant difference between treatments (95% CI [-14.87, 11.21], d = 0.16). Conclusions: Acceptance-based procedures may be a viable alternative to traditional CBT techniques in the management of tinnitus. The Internet can improve access to psychological interventions for tinnitus. (Contains 2 figures, 2 tables, and 4 footnotes.)