The Organization for Economic Cooperation and Development (OECD) Main Steam Line Break (MSLB) Benchmark was defined to validate the thermal-hydraulic system codes coupled with three-dimensional (3-D) neutron kinetic codes. The reference problem is an MSLB in a pressurized water reactor at end of cycle. The analyses were performed with the 3-D core model DYN3D, the thermal-hydraulic system code ATHLET, and the coupled code DYN3D/ATHLET. The results of the DYN3D and ATHLET simulations based on the specification are compared with the results of other participants in the final OECD reports. The effect of the thermal-hydraulic nodalization of the core, i.e., the number of coolant channels, and the influence of the coolant mixing inside the pressure vessel are studied in the paper. Calculations with a reduced number of coolant channels are performed often in coupled calculations for saving computational time. Results of a 25-channel model were compared with the 177-channel calculation (1 channel per assembly). The results for global parameters like nuclear power show only small differences for the two models; however, the prediction of local parameters such as maximum fuel temperatures requires a detailed thermal-hydraulic modeling. The effect of different coolant mixing within the reactor pressure vessel is investigated. It is shown that the influence of coolant mixing mitigates the accident consequences when 3-D neutron kinetics is applied. In case of point kinetics, coolant mixing leads to an opposite effect. To profit from the 3-D core model, a realistic description of the coolant mixing in the coupled codes is a topic of further investigations.