The goal of this study was to determine the ability of the Hanover gilt model to assess in vivo fertilizing capacity of preserved sperm and to consider whether any modifications to this model were needed. This model evaluates the fertilizing capacity of semen based on the fertilization rate, the rate of normal embryos and the accessory sperm count of 3-5-day embryos. Its distinguishing characteristics are the use of one-time insemination of sperm in reduced numbers, of spontaneously ovulating gilts and of ovulation detection through ultrasound examination of ovaries. Reduced sperm numbers allow for an accurate evaluation of the fertilizing potential of different semen treatments, thereby avoiding the compensatory effect of doses calibrated to maximize fertility. The model's usefulness was assessed in a trial run designed to compare the fertilizing capacity of liquid boar semen diluted into two different extenders. The diluent, the boar and the backflow, had no significant effect on any of the parameters studied. Gilts inseminated less than 24 h before ovulation had a significantly higher (p < 0.01) fertilization rate and accessory sperm cell count (p < 0.05) than those inseminated more than 24 h before ovulation. Very good/good embryos from homogeneous litters (only very good/good embryos were present) had a significantly higher (p < 0.01) accessory sperm count than those from heterogeneous litters (at least one embryo was of a different quality and/or oocytes were present). Both very good/good and degenerated/retarded embryos from heterogeneous litters had low accessory sperm numbers. This suggests that accessory sperm count is significantly related to the quality of the litter, but not to the quality of the embryo within gilts. It can be concluded that the Hanover gilt model is sensitive enough to show fertility differences (in this study, those associated with in vivo ageing of semen), while using relatively few gilts and little time.