Objective: Cone beam computed tomography (CBCT) images can be useful for estimating cervical vertebrae maturity (CVM). The aim of this study was to evaluate the reliability of cephalograms derived from CBCT versus lateral cephalograms (LC) to estimate the CVM in a Peruvian population. Material and methods: The sample evaluated consisted of 40 cephalograms derived from CBCT and 40 LC images from individuals aged 10–19 years. One trained and calibrated observer (Kappa scores ≥ 0.90) interpreted the CBCT and LC images twice. Intra-observer reliability of each maturation stage on CBCT sagittal slices and LC images were analysed using the weighted kappa statistics (α = 0.05). Comparison of CVM stages between CBCT slices and LC images were analysed by the Spearman rank correlation coefficient, p < 0.05. Results: The weighted kappa test showed almost perfect intra-observer agreement for the CVM stages using the CBCT sagittal slices (0.873). Considering the LC images, the weighted kappa test showed almost perfect intra-observer agreement too (0.937). In both intra-observer agreement, the difference was limited to one maturation stage of the CVM method. The first and second intra-observer agreement for the CVM stages between the CBCT sagittal slices and LC images were almost perfect (0.937 and 0.874). High correlation values at the first (0.975) and second (0.976) intra-observer agreement for the CVM stages between CBCT sagittal slices and LC images were also found. Conclusion: CBCT is a reliable method for CVM assessment and can be used as an alternative method for this purpose. The orthodontists might use the CBCT scans as a valuable tool for CVM method estimation.