Internet-delivered cognitive behavior therapy (ICBT) has been tested in many research trials, but to a lesser extent directly compared to face-to-face delivered cognitive behavior therapy (CBT). We conducted a systematic review and meta-analysis of trials in which guided ICBT was directly compared to face-to-face CBT. Studies on psychiatric and somatic conditions were included. Systematic searches resulted in 13 studies (total N=1053) that met all criteria and were included in the review. There were three studies on social anxiety disorder, three on panic disorder, two on depressive symptoms, two on body dissatisfaction, one on tinnitus, one on male sexual dysfunction, and one on spider phobia. Face-to-face CBT was either in the individual format (n=6) or in the group format (n=7). We also assessed quality and risk of bias. Results showed a pooled effect size (Hedges' g) at post-treatment of -0.01 (95% CI: -0.13 to 0.12), indicating that guided ICBT and face-to-face treatment produce equivalent overall effects. Study quality did not affect outcomes. While the overall results indicate equivalence, there are still few studies for each psychiatric and somatic condition and many conditions for which guided ICBT has not been compared to face-to-face treatment. Thus, more research is needed to establish equivalence of the two treatment formats.