The purpose of study was to determine accuracy of 2-dimensional parameters and 3-dimensional surface distance deviation of dental models derived from two intraoral scanners compared to plaster models taken from samples with malocclusion and samples who previously received orthodontic treatment. The samples consisted of 15 subjects with malocclusion (PAR scores > 24, Malocclusion Group) and 15 subjects who previously received orthodontic treatment (PAR scores < 9, Orthodontic treated Group). Three types of dental models from each subject were (1) plaster models from Alginate impression, (2) digital models from intraoral scanners, TRIOS® (3Shape A/S, Copenhagen, Denmark) and (3) digital models from CEREC Ortho® (Sirona, Dental, Thailand). Nine 2-D parameters of plaster models using a digital caliper were measured. The digital models were measured by digitization processing software (MeshLab, Italy). Surface distance deviation of digital models from the two scanners were determined by superimposing surfaces using color mapping. Comparing 2D parameters by RM-ANOVA with Bonferroni’s method for multiple comparisons, results showed that Upper Intercanine Width of plaster models was significantly different from CEREC Ortho® digital models (F(2,56) = 4.490, p= .016) with mean difference -0.146±0.057 mm (p= .048). In addition, malocclusion had a significant interaction on Midline Deviation (MD) among three types of models (F(1.408, 39.430) = 5.421, p= .016). The plaster models had significant difference MD from digital models from TRIOS® and CEREC Ortho® (Mean difference 0.234±0.068 mm, p= .005; 0.169±0.066 mm, p= .047 respectively). Comparing by t-test, there was no significant difference of 3-D surface distance deviation of digital models between Malocclusion and Orthodontic treated groups. In conclusion, accuracy of two digital models was clinically accepted. Malocclusion had an interaction on Midline Deviation of the two digital models. 3-D surface distance deviation of the digital models from two intraoral scanners showed no significant difference.