In this study, the accuracy of structured and unstructured mesh CFD codes in simulating the flow around a threeelement high-lift configuration (slat, main wing, and flap) is assessed, and mesh dependency and effect of turbulence models are studied. In the first part of the study, the results of two structured mesh CFD codes and an unstructured mesh CFD code using the same turbulence model are compared and discussed. A mesh refinement approach is also used to examine the dependency of numerical accuracy on the density of unstructured meshes. By properly distributing mesh points in an unstructured mesh, the detail in flow physics is obtained. It is also shown that the quantitative prediction of the vortex in the slat cove is an important factor that affects accuracy. In the second part of the study, three turbulence models are compared using one of the structured mesh CFD codes. All turbulence models can produce similar flowfields. However, several differences are seen in the separated regions, especially in the slat cove. Commonly used turbulence models, Spalart-Allmaras model and Menter's SST model, produce similar aerodynamic forces at lower angles of attack. At higher angles of attack, the SST model gives better results for the present computations. It is found that the maximum lift and the angle at which the stall occurs are very sensitive to the turbulence model.