Orthodontists evaluate the cephalometric analysis in order to determine a correct diagnosis. In clinical practice, oral radiologists are the group of specialists who run this test. However, due to several factors, some orthodontists feel the need to confirm the results reported by oral radiology clinics and perform the cephalometric analysis again. The main objective of this study was to assess the consistency among cephalometric measurements obtained by orthodontists and radiologists using computerized cephalometric analysis software.Thirty orthodontists and 30 oral radiologists identified 18 cephalometric landmarks using the same computer, as directed by the software Radiocef Studio 2®. From there, 14 cephalometric parameters were generated. In order to verify the intraexaminer agreement, 10 professionals from each group repeated the identification of the landmarks with a minimum interval of 8 days between the two markings. The intragroup variability was calculated based on the coefficients of variation. The comparison between groups was performed by using the Student t-test and the Mann–Whitney test.In the group of orthodontists, the measurements of Pog and 1-NB, SL, line S-Ls, Line S-Li, and 1.NB showed high internal variability. In the group of radiologists, the same occurred with the values of Pog and 1-NB, line S-Ls, Line S-Li, and 1.NA. In the comparison between groups, all the analyzed linear measurements (Pog and 1-NB, SL, Line S-Ls, and Line S-Li) and two angular measurements (1.SN and AF) showed statistically significant differences between radiologists and orthodontists (P < 0.05).All linear measures assessed (and only two angular measures) indicated an inconsistency between orthodontists and radiologists.