This study compared the dimensional changes between computer-aided design and computer-aided manufacturing (CAD-CAM) milled complete denture bases (CDBs) and three-dimensional (3D) printed CDBs. Materials and methods: One maxillary completely edentulous stone model was fabricated with three reference points at the incisive papilla, right molar, and left molar areas marked as X, Y, and Z, respectively. It was scanned to produce a standard tessellation language (STL) file, which was imported to a metal milling machine software to produce the metal model. This metal model was used to fabricate 30 CDBs for analysis. The CDBs were divided into three groups (n = 10 each) according to the fabrication method used as follows: Group 1, CAD-CAM milled CDBs; Group 2, 3D printed CDBs; and Group 3, conventional compression molded CDBs. The CDBs of all groups were scanned after fabrication, and the dimensional changes in each were evaluated by two methods. The first was the two-dimensional evaluation method that involved linear measurement of the distances between the reference points (X-Y, X-Z, and Y-Z) of the scanned reference cast and dentures. The second method was the 3D evaluation method that involved the superimposition of the STL files of the dentures on the STL file of the reference cast. Data were calculated and were statistically analyzed using one-way analysis of variance and Tukey's pairwise post hoc tests. Results: There was a significant difference in the dimensional accuracy between the CAD-CAM milled, 3D printed, and conventional compression molded CDBs (p < 0.05). Conclusion:The dimensional accuracy of the CAD-CAM milling system in complete denture fabrication is superior to that of the compression molding and 3D printing systems.