During the last decade, biological therapies have an increasing share in the modern therapeutics of various diseases including inflammatory bowel diseases (IBD). Animal models of IBD have often been used to identify the targets of biological therapies, to test their relevance to disease pathogenesis, to assess their therapeutic efficacy in vivo, and to check for drug toxicity. In the field of inflammatory diseases the majority of biologics under development have failed to reach the clinic. This review examines the ability of preclinical data from animal models of IBD to predict success or failure of biologics in human IBD. Specifically, it describes the murine models of IBD, the mechanism of disease induction, the phenotype of the disease, its relevance to human IBD, and the specific immunological features of disease pathogenesis in each model and mainly compares the results of the phase II and III trials of biologics in IBD with preclinical data obtained from studies in animal models. Finally, it examines the possible reasons for low success in translation from bench to bedside and offers some suggestions to improve translation rates.