A direct, simultaneous comparison of bone substitutes is hampered by the limited number of samples that can be tested simultaneously. The goal of this study was to establish a preclinical model for guided bone regeneration that offers testing of different bone substitutes in a one-wall defect situation. We show here that up to eight titanium hemispheres can be placed on the calvaria of minipigs. To establish our model, titanium hemispheres were filled with and without Bio-Oss, a deproteinized bovine bone mineral, Ostim, an aqueous paste of synthetic nanoparticular hydroxyapatite, and Osteoinductal, an oily calcium hydroxide suspension, before being positioned on the calvaria. After 6 and 12 weeks, titanium hemispheres were subjected to histological and histomorphometric analysis. We show here that bone filled approximately one-tenth of the area below the hemispheres which were left empty, indicating a critical size model for guided bone regeneration. In accordance with the documented osteoconductive properties of Bio-Oss and Ostim, titanium hemispheres were almost completely filled with bone. Moreover, the expected degradation profile of Bio-Oss and Ostim could be confirmed by histologic and histomorphometric analysis. Under the same conditions, Osteoinductal failed to exert osteoconductive properties, rather a progressive resorption of the host bone was observed. These results demonstrate that the preclinical model presented here is suitable to simultaneously compare bone substitutes with different material properties. Our model based on the titanium hemispheres allows evaluation of graft consolidation under standardized conditions thereby avoiding intra-individual variations.