The Weibull distribution is the most widely used function in the reliability analysis and structural design of dental ceramics; however, it is still unclear whether Weibull distribution is always the most suitable one. With wide applications of dental ceramics, a special attention has been paid in discriminating their strength distributions. In this paper, three versatile functions, involving normal, log-normal and Weibull distributions, are applied to the analysis of ten strength data sets of dental ceramics with different compositions and the results are compared in terms of the Akaike information criterion and the Anderson–Darling test. It reveals that various microstructures and compositions in the investigated dental ceramics cause their strength distributions deviated from the Weibull distribution. The influence of microstructure induced fracture properties (multiple-modal flaw size distribution, R-curve behavior and subcritical crack growth) on strength distributions is discussed.