The generalized deterministic vertex model was successfully used to study the recrystallization process and the corresponding results were published elsewhere . In its classical form the vertex model has analytical formulation, basing on the total energy (i.e. boundary energy and stored energy) minimization. A change of grain boundary configuration in classical vertex model is found by the calculation of vertex velocities. Consequently, a global and complex system of equations has to be solved in each step. In order to simplify calculations and to handle the problem in a more flexible way, the statistical model was proposed. Typical elements of Monte Carlo algorithm were incorporated into the vertex model: a random (and small) modification of microstructure is accepted with the probability proportional to Boltzmann factor. This approach is closer to the stochastic nature of recrystallization process. The model was used to study the recrystallization of 70% and 90% cold rolled polycrystalline copper. It predicts correctly recrystallization textures for high and low strains.