Objective Tissue-engineered human oral mucosal models have been developed for biocompatibility assessment of biomaterials. The aim of this study was to evaluate the biological effects of three different composite resin systems on a three-dimensional human oral mucosal model. Methods Full-thickness oral mucosal models were engineered by air/liquid interface culture of a human oral keratinocyte cell line on a lamina propria composed of oral fibroblasts seeded into a porous scaffold. The surface of the tissue models was exposed to three types of experimental composite resins: a TEGDMA-based, a UDMA-based, and a BisGMA/TEGDMA (80:20)-based composite resin for 24 h. The response of the engineered oral mucosa to the test materials was assessed using routine histology, the Alamar Blue tissue viability assay and IL-1β release measured by ELISA. Results Compared to the other materials tested, the TEGDMA-based composite resin caused significant damage to the oral mucosal model. Statistical analysis by one-way ANOVA followed by Tukey's analysis showed that there was a significant decrease in the viability of tissue models after 24 h exposure to TEGDMA-based composite resin. Also exposure to TEGDMA-based composite resin significantly increased the amount of IL-1β released from the oral mucosal model. Conclusion The 3D human oral mucosal model has the potential to be a more relevant and more informative model than monolayer cell culture systems for biocompatibility testing of dental materials. The results obtained from multiple-endpoint analysis of the oral mucosal model indicate significant mucotoxicity of high TEGDMA-containing composite resins.