The primary objective of this study was to assess the performance of different genomic prediction models applied to the selection of purebreds for crossbred performancebased on high-density marker data. Our results suggest that in the presence of dominant gene action, selection based on the dominance model is superior to both the a breed-specific allele model and an additive model in terms of maximizing crossbred performance through purebred selection, especially when training is not updated each generation