This paper presents probabilistic assessment of load-bearing capacity and reliability for different STM of beams loaded with a torsional and bending moment. Three beams having different reinforcement arrangement obtained on the basis of STM but the same overall geometry and loading pattern were analysed. Stochastic modelling of this beams were performed in order to assess probabilistic load-bearing capacity. In the analysis, the random character of input data - concrete and steel was assumed. During the randomization of variables the Monte Carlo simulation with the reduce the number of simulations the Latin Hypercube Sampling (LHS) method was applied. The use of simulation methods allows for approximation of implicit response functions for complex in description and non-linear reinforced concrete structures. On the basis of the analyses and examples presented in the paper, it has been shown that the adoption of different ST models determines the different reliability of the analysed systems and elements.