The topic of this work is the numerical simulation of a turbulent nonpremixed hydrogen (H-2) jet flame with different combustion models. The predictions are validated against existing experimental data provided by Raman and laser Doppler anemometry (LDA) measurements for a turbulent jet hydrogen-air diffusion flame [1, 2]. In particular, a comparison of two "advanced" turbulent combustion models is presented: These are a probabilistic Euler Lagrangian (PEUL) model based on the idea of Lagrange interaction by exchange with the mean (IEM), and a model based on the scalar probability density function (PDF) transport equation for the thermochemical variables, which is solved using a Monte Carlo method. In addition, a "standard" Eddy Dissipation Model with a single-step reaction is considered. The numerical results for mean velocity components, turbulent kinetic energy, mixture fraction, temperature, and major chemical species are presented and compared with the experimental data. The goal of the work is to investigate the capabilities of the used models in predicting hydrogen combustion in a jet flame. This simple geometry allows for reliable flow simulations. Regarding the basic test case under consideration, the results obtained by the PEUL computations and the Monte Carlo simulation are in good agreement with experimental data. The comparison shows that both probabilistic methods give better predictions than the "standard" model. The advantages and disadvantages of the models are discussed in detail in relationship to the results. It is possible to draw conclusions for modeling improvements. The improved models should be able to be applied also to more complex geometries.