Objective:To investigate the accuracy of pure titanium and cobalt-chromium alloy frameworks fabricated using the additive manufacturing (AM) of selective laser melting technology (SLM) for the mandibular implant-supported fixed prostheses and the maxillary removable partial denture (RPD), and to provide a reference for clinical application of SLM pure titanium frameworks.Methods:One edentulous mandibular model with implants and screw fixed abutments at bilateral canines and the first molars was selected and used as the mandibular full arch implant-supported model. At the same time, a Kennedy class Ⅰ maxillary dentition defect model was selected. The digital models were obtained by scanning the dental models, and the metal frameworks of the mandibular full arch implant-supported denture and the maxillary RPD (design model) were designed using the 3 Shape software. Meanwhile, 12 mandibular frameworks in the cobalt-chromium alloy and the pure titanium (6 in each group were treated with heat treatment, while the other 6 were not treated), and 7 maxillary frameworks in the cobalt-chromium alloy and the pure titanium were respectively made by SLM with the improved dual-laser metal printer. The axial direction of the printing powder accumulation was taken as the Z-axis. During the design process, the software (3Shape Dental System 2018) automatically generated the X-axis and Y-axis, X axis was the sagittal axis of the dental model and Y axis was the coronal axis of the dental model. The deviation of the interface center of the abutment of the digital model of the mandibular frameworks from the design model in the X, Y and Z axes was analyzed. As for the trueness of the mandibular framework, the larger the deviation data was, the worse the trueness was. The deviation of the whole maxillary framework and 7 measuring points (palatal plate center point and bilateral occlusal rests, I bars, proximal plates) were analyzed. The fitness of the whole maxillary framework to the design model was expressed by root mean square (RMS) of the deviation data, and the fitness of measuring points was expressed by the mean±standard deviation of the data. The trueness differences of each group before and after heat treatment of the mandibular framework and the fitness of the maxillary framework were compared.Results:The cobalt-chromium alloy frameworks showed lower trueness on the X, Y, Z-axes [(96.3±12.1), (86.3±11.4), (61.2±13.2) μm] than did the pure Ti frameworks [(82.3±11.2), (72.2±10.2), (51.2±11.6) μm] by SLM, and the heat treatment could reduce the discrepancy between the SLM frameworks and STL models, for pure titanium frameworks [(62.4±11.3), (55.2±13.2), (41.3±10.8) μm] and for cobalt-chromium alloy [(84.5±10.5), (72.3±11.2), (54.2±11.6) μm]. For the thin RPD major frameworks, pure titanium had better fitness [(121.3±17.0) μm] than cobalt-chromium alloy [(174.0±18.3) μm] by SLM, and the difference was statistically significant ( P<0.05). Conclusions:Pure titanium frameworks fabricated by SLM additive manufacturing technology exhibited better fitness and trueness than did the Co-Cr frameworks after heat treatment respectively, and this satisfied the requirements of implant-supported fixed prostheses and RPD major metal frameworks.