Fabrication of complete dentures using a 3D printer is quicker and more economic than conventional methods. However, the photopolymer resins used in 3D printers has a lower flexural strength than heat-cured resin. Furthermore, photopolymer resins exhibit anisotropic properties depending on the printing direction, but no studies have evaluated their mechanical properties. The impact of stress distribution caused by changing the printing direction of the 3D printed denture has not been clarified. This study aimed to investigate the effect of different printing directions (0 degrees, 45 degrees, and 90 degrees) of stereolithography (SLA) 3D printed dentures on stress distribution. Artificial mucosa was fabricated to fit a maxillary edentulous model, which was scanned to generate a standard tessellation language (STL) file. Subsequently, the upper denture was designed using computer-aided design (CAD) software, output as an STL file (master data), and set in three different printing directions (0 degrees, 45 degrees, and 90 degrees). It was printed by the SLA 3D printer using photopolymer resin (n=6, in each printing direction). The stress distributions of the dentures were monitored using four rosette strain gauges, which were cemented to the midline of each denture as follows: above the labial frenum (A), at the incisive papilla (B), at the endpoint of the denture (D), and at the mid-point of B and D (C). A load was applied to the posterior region at a loading rate of 20 N/s from 0 N to 200 N using a universal testing machine. Changes in the applied load and strain at each point were recorded. The maximum principal strain (MPS) and the direction of the MPS (theta) were calculated. Each mean MPS was compared using Kruskal-Wallis and Steel-Dwass multiple comparison tests (p < 0.05). The stress distribution showed that the MPS for 45 degrees printing was the lowest at each measurement point except for A under 200 N loading. The MPS at C for 0 degrees was significantly different from that at C for 45 degrees and 90 degrees. The MPS at D in all the printing directions showed significant differences. The MPS at B for 45 degrees was significantly different from that at B for 90 degrees. The MPS at A in all printing directions showed no significant difference. Within the limitations of this study, differences in printing direction affected the stress distribution of SLA 3D printed dentures. The results showed that the stress distribution of the denture printed at 45 degrees by the 3D printer system was the smallest compared with dentures printed at 0 degrees and 90 degrees. These results suggest that a printing direction of 45 degrees is preferable when fabricating dentures using a 3D printer in clinical setting.