Objectives: The accuracy of computer-aided design/computer- aided manufacturing (CAD/CAM) systems is linked to their technical characteristics and reliability for manufacturing the restoration designed. The aim of this study was to compare the accuracy of fit of zirconia copings manufactured with different CAD/CAM systems and their capacity to conform to pre-established spacer design settings. Methods: Sixty zirconia copings were manufactured by three CAD/CAM systems, with their spacers set as recommended by their manufacturer on occlusal, axial, and marginal surfaces. The 3D triple-scan optical technique was used to obtain a fit mapping and to analyze the marginal and axial accuracy of fit. The reliability of the 3D measurement method was estimated using intraclass correlation and repeatability coefficients. The preparation coping interface width results were statistically analyzed using non-parametric analysis (Kruskal-Wallis, one-way ANOVA, and Wilcoxon signedrank tests) (P 0.05). Results: The repeatability coefficient was 6, 8, and 15 μm for axial, marginal, and occlusal interface width measurements, respectively. For the three systems tested, no differences were found in the marginal area of the copings studied, with a mean fitting accuracy ranging from 54.3 to 66.6 μm interface width. Statistically significant differences between groups were observed for the fitting accuracy measured in axial and occlusal areas. With the spacers set in the different areas, mean fit measurements of the zirconia copings were significantly larger, with an increased fit width ranging from 30 to 73 μm. Significance: The three CAD/CAM systems evaluated allowed similar marginal accuracy but failed to reproduce the pre-established spacer parameters, with larger spacing showing throughout. Palabras clave: 3D accuracy of evaluation, accuracy of fit, CAD/CAM, dental prostheses, design parameters