Objective: This research aimed to evaluate the effect of different 3D printing layer thicknesses (50-μm and 100-μm) on the accuracy of implant surgical templates fabricated by a digital light processing (DLP) printer. Material and methods: The Nissin educational model was scanned three times (missing maxillary central incisor, missing maxillary canine, missing maxillary first molar). Each time will be scanned individually to generate a virtual STL model, which model will be imported into the planning software and virtual implant planning is completed and the virtual implant guide will be generated, and 3D printed using DLP 3D printer, the guides are printed in 2 different vertical resolutions (50-μm and 100-μm). Then all the printed implant surgical guides were scanned at an interval one week after printing, the scan STLs files will be superimposed to each design file to detect surface deviations from the original design file by Geomagic reverse engineering software. Result: Data showed significant deviation in the first molar group for the 100-μm layer compared to the 50-μm layer (p=0.002). In the central incisor and canine groups, deviations were higher for the 100-μm layer but not statistically significant (p=0.611 and p=0.176, respectively). Conclusion: Results show that implant surgical guides printed at 50-μm display overall lower deviations and smoother surface when compared to 100-μm implant surgical guides in both the fitting surface and the guide tube.