Disproportionate dwarfism phenotypes represent a heterogeneous subset of skeletal dysplasias and have been described in many species including humans and dogs. In this study, we investigated Vizsla dogs that were affected by disproportionate dwarfism that we propose to designate as skeletal dysplasia 3 (SD3). The most striking skeletal changes comprised a marked shortening and deformation of the humerus and femur. An extended pedigree with six affected dogs suggested autosomal recessive inheritance. Combined linkage and homozygosity mapping localized a potential genetic defect to a ~4 Mb interval on chromosome 33. We sequenced the genome of an affected dog, and comparison with 926 control genomes revealed a single, private protein-changing variant in the critical interval, PCYT1A:XM_038583131.1:c.673T>C, predicted to cause an exchange of a highly conserved amino acid, XP_038439059.1:p.(Y225H). We observed perfect co-segregation of the genotypes with the phenotype in the studied family. When genotyping additional Vizslas, we encountered a single dog with disproportionate dwarfism that did not carry the mutant PCYT1A allele, which we hypothesize was due to heterogeneity. In the remaining 130 dogs, we observed perfect genotype–phenotype association, and none of the unaffected dogs were homozygous for the mutant PCYT1A allele. PCYT1A loss-of-function variants cause spondylometaphyseal dysplasia with cone–rod dystrophy (SMD-CRD) in humans. The skeletal changes in Vizslas were comparable to human patients. So far, no ocular phenotype has been recognized in dwarf Vizslas. We propose the PCYT1A missense variant as a candidate causative variant for SD3. Our data facilitate genetic testing of Vizslas to prevent the unintentional breeding of further affected puppies.