The aim of this study was to analyze the reliability and repeatability of identification landmarks using 3-D cephalometric software. Ten orthognathic patients were selected for this study and underwent the following protocol: 1. radiographic evaluation (CBCT technique); 2. stone casts; 3. photos; and 4. 3-D cephalometric evaluation. Twenty-one hard tissue landmarks and 14 cephalometric measurements were taken three times (T1, T2, and T3) on each patient, with an interval of one week by two experts in orthodontics (A, B). Standard deviation and Pearson's correlation coefficient were calculated to evaluate intra- and inter-observer repeatability. The results showed a strong correlation for both intra- and inter-observer Pearson's correlation coefficient (>0.7). The current preliminary study showed that the reliability and repeatability of the identification landmarks were very high if the 3-D cephalometric landmarks are defined correctly in the three planes of the space. Further evaluation is necessary to better define the 3-D cephalometric system.