Mate choice for genetic benefits remains controversial, largely because few studies have estimated the relative contributions of additive and non-additive sources of genetic variation to offspring fitness. Moreover, there remains a deficit of these estimates for species where female-mate preferences have been quantified in the wild, especially species characterized by monandry or monogamy. Here, we use artificial fertilization techniques combined with a cross-classified breeding design to simultaneously test for "good genes" and "compatible genes" benefits of mate choice in the monandrous red backed toadlet (Pseudophryne coriacea). In addition, we used a genomic approach to estimate effects of parental-genetic relatedness (assessed using 27, 768 single nucleotide polymorphisms) on offspring fitness. Our results revealed no significant additive genetic effects (sire effects), but highly significant non-additive genetic effects (sire × dam interaction effects), on fertilization success, survival during embryonic development, and hatching success. We also found significant associations between parental genetic similarity and offspring survival (whereby survival was higher when parents were more related), and significant positive relationships between fertilization success and embryo survival through to hatching. These results indicate that offspring viability is significantly influenced by the genetic compatibility of parental genotypes, that more related parents are more genetically compatible, and that gametes with greater compatibility at fertilization produce more viable offspring. More broadly, our findings provide new quantitative genetic evidence that genetic incompatibility underpins female mate preferences. Continued quantitative genetic assessment of the relative importance of good genes versus compatible genes is needed to ascertain the general importance of genetic benefits as a driver of female mate choice.