In case of large segmental defects in load-bearing bones, an external fixator is used to provide mechanical stability to the defect site. The overall stiffness of the bone–fixator system is determined not only by the fixator design but also by the way the fixator is mounted to the bone. This stiffness is an important factor as it will influence the biomechanical environment to which tissue engineering scaffolds and regenerating tissues are exposed. A finite element (FE) model can be used to predict the system stiffness. The goal of this study is to develop and validate a 3D anatomical FE model of a bone–fixator system which includes a previously developed unilateral external fixator for a large segmental defect model in the rabbit tibia. It was hypothesized that the contact interfaces between bone and fixator screws play a major role for the prediction of the stiffness. In vitro mechanical testing was performed in order to measure the axial stiffness of cortical bone from mid-shaft rabbit tibiae and of the tibia–fixator system, as well as the bending stiffness of individual fixator screws, inserted in bone. μCT-based case-specific FE models of cortical bone and SCREW–BONE specimens were created to simulate the corresponding mechanical test set-ups. The Young's modulus of rabbit cortical bone as well as appropriate screw–bone contact settings were derived from those FE models. We then used the derived settings in an FE model of the tibia–fixator system. The difference between the FE predicted and measured axial stiffness of the tibia–fixator system was reduced from 117.93% to 7.85% by applying appropriate screw–bone contact settings. In conclusion, this study shows the importance of screw–bone contact settings for an accurate fixator stiffness prediction. The validated FE model can further be used as a tool for virtual mechanical testing in the design phase of new tissue engineering scaffolds and/or novel patient-specific external fixation devices.