This work investigates the turbulent flow in the high head Francis turbine of the Tokke model. The geometry and measurement data of the herein considered Tokke scaled down model are published on the Francis 99 workshop homepage . Based on this data steady-state and unsteady flow simulations are set up in commercial code and unsteady flow simulations are set up in OpenFOAM at the BEP (best efficiency point) condition. The whole domain is meshed in Ansys ICEM CFD using a block structured grid and grid refinement towards the hydraulic smooth walls. The wall boundary layer is resolved by y+mean 30$ and automatic wall functions. Turbulent flow behavior is modeled by applying the shear stress transport model, komega-SST. The grid convergence method verifies the ability of monotonic convergence. All further computations are based on this grid type meeting computational efficiency and validity of the calculated quantities. In contrast to the one passage model setup of the grid error estimation, the further simulations are performed on the full turbine model with 5M cells. Commercial code: The efficiency is predicted accurately in the BEP and HL (high load) condition in the steady-state and unsteady simulation, with a maximum total deviation of eta% < 1.5%. In the PL (part load) regime no obvious trend emerges for the global quantities. The averaged static pressure at the probes is estimated in accordance with the literature , , . The tendency of underpredicting the probes in the draft tube and overpredicting the pressure in the runner domain, holds for all operating conditions. The computed mean velocity profiles coincide for the steady-state and unsteady CFD (computational fluid dynamics) simulation. In PL the velocity is captured accurately, whereas in the two other operating points, local effects are predicted incorrectly. However, the simulation results are consistent with , . The main features of the pressure fluctuations are captured in the vaneless space. The efficiency losses in the distributor and the draft tube are evaluated according to  in all three operating points. OpenFoam: is not capable of predicting the global quantities compared to the experimental data and commercial code. The efficiency deviates by eta% = -13.2%. However, the local static pressure data is estimated similar to commercial code.