This study sought to evaluate and compare the marginal and internal fit in vitro of three-unit FDPs in Co-Cr made using four fabrication techniques, and to conclude in which area the largest misfit is present. METHODS: An epoxy resin master model was produced. The impression was first made with silicone, and master and working models were then produced. A total of 32 three-unit Co-Cr FDPs were fabricated with four different production techniques: conventional lost-wax method (LW), milled wax with lost-wax method (MW), milled Co-Cr (MC), and direct laser metal sintering (DLMS). Each of the four groups consisted of eight FDPs (test groups). The FDPs were cemented on their cast and standardised-sectioned. The cement film thickness of the marginal and internal gaps was measured in a stereomicroscope, digital photos were taken at 12× magnification and then analyzed using measurement software. Statistical analyses were performed with one-way ANOVA and Tukey’s test. RESULTS: Best fit based on the means (SDs) in μm for all measurement points was in the DLMS group 84 (60) followed by MW 117 (89), LW 133 (89) and MC 166 (135). Significant differences were present between MC and DLMS (p<0.05). The regression analyses presented differences within the parameters: production technique, tooth size, position and measurement point (p<0.05). SIGNIFICANCE: Best fit was found in the DLMS group followed by MW, LW and MC. In all four groups, best fit in both abutments was along the axial walls and in the deepest part of the chamfer preparation. The greatest misfit was present occlusally in all specimens.