Purpose: The aim of this study was to determine the reliability and accuracy of measurements in digital models(CEREC$^{(R)}$ AC) compared to stone models. Methods: A master model(500B-1, Nissin Dental Product, Japan) with the prepared upper full arch tooth was used. Conventional impression and then stone model(n=10) were produced from this master model, and on the other hands, digital impressions were made with the CEREC$^{(R)}$ AC intra-oral scanner(n=10). One examiner measured two times the intercanine, intermolar distance, dental arch length. The stone model were measured using a digital caliper. The t-student test for paired samples and intraclass correlation coefficient(ICC) were used for statistical analysis. Results: The measurement of two methods showed very good reliability. At the intra-examiner reliability of measurement, ICC at the stone and CEREC$^{(R)}$ AC model were 0.81 and 0.94. The mean difference between measurements made directly on the stone models and those made on the CEREC$^{(R)}$ AC model was 0.20~0.28mm, and was statistically significant(P=0.001). Conclusion: These in vitro studies show that accuracy of the digital impression is similar to that of the conventional impression. These results will have to be confirmed in further clinical studies.