Aimed to compare the volumetric calculations (VC) of intra-and-extraoral scanners on carious teeth before/after caries removal. 120 extracted human molars with ICDAS scores of 3, 4, and 5 were included. The teeth were scanned using an extraoral scanner (Ineos-X5, Dentsply Sirona) and three intraoral scanners (IOS) (iTero Element-5D, Align Technology; Primescan, Dentsply Sirona; Trios 4, 3Shape) before-and-after caries removal (CR). Eight Standard Tessellation Language (STL) data of each tooth sample were overlapped in Meshmixer (Autodesk) software for VC. Shapiro-Wilk, Paired Two-Sample T-test, repeated analysis of variance test statistics, and intraclass correlation coefficient (ICC) were used (p < 0.05). In initial VC, a significant difference observed between Ineos-X5 and iTero Element-5D (p < 0.001) and Ineos-X5 and Primescan (p < 0.001), regardless of the ICDAS score. No significant difference observed after CR between Ineos-X5 and iTero Element-5D (p = 0.917), Ineos-X5 and Primescan (p = 0.936), and Ineos-X5 and Trios 4 (p = 0.727) respectively. There was also no difference between the three IOS after CR (p ≥ 0.05), whereas the iTero Element-5D scanner significantly calculated less volume than Primescan and Trios-4 initially (p = 0.003). Maximum ICC was observed between the Ineos-X5 and Primescan in the ICDAS 5 score, before CR (ICC = 1, p < 0.001) and triple comparisons of iTero Element-5D, Primescan, and Trios-4 (ICC = 1, p < 0.001). Carious and irregular surfaces might affect the data collection of IOS. Intra-and-extraoral data may present negligible differences in the volumetric calculation, depending on the caries status and anatomical properties. These differences and factors may be important for future artificial intelligence networks that simulate the aftermath of caries removal. The IOSs tested in the current study can successfully collect data from irregular, deep and shallow cavities after caries removal.