This paper presents a probabilistic framework for the modeling of fatigue damage in three gamma-based titanium aluminides: Ti-45Al-2Mn-2Nb+0.8 vol% TiB 2 , Ti-47Al-2Mn-2Nb+0.8 vol% TiB 2 and Ti-48Al-2Cr-2Nb (all compositions quoted in atomic percent unless stated otherwise). This includes: empirical stress life and fracture mechanics approaches to the estimation of material reliability or the risk of failure. Empirical reliability functions are obtained initially from multiple stress-life experiments designed to identify the statistical distributions that best describe the measured variabilities in fatigue life. Fracture mechanics-based reliability functions are also derived using statistical distributions that characterize the measured variabilities in fatigue crack growth data obtained from fatigue crack growth experiments in the long crack regime. The anomalous behavior of short cracks is then discussed before assessing the implications of the current work for the design of engineering structures and components from gamma-based titanium aluminides.