The dog is valued as a companion animal and increasingly recognized as a model for human disorders. Given the importance of T cells in health and disease, comprehensive knowledge of canine T cells can contribute to our understanding of pathogenesis mechanisms and inform the development of new treatment strategies. However, the diversity of canine T cells is still poorly understood mainly due to the lack of species-reactive antibodies for use in flow cytometry. The aim of this study was to generate a detailed atlas of peripheral blood TCRαβ⁺ T cells of healthy dogs using single-cell RNA-sequencing (scRNAseq) combined with immune repertoire sequencing. A total of 22 TCRαβ⁺ T cell clusters were identified, which were classified into three major groups: CD4-dominant (11 clusters), CD8A-dominant (8 clusters), and CD4/CD8A-mixed (3 clusters). Based on differential gene expression, distinct differentiation states (naïve, effector, memory, exhausted) and lineages (e.g. CD4 T helper and regulatory T cells) could be distinguished. Importantly, several T cell populations were identified, which have not been described in dogs before. Of particular note, our data provide first evidence for the existence of canine mucosa-associated invariant T cell (MAIT)-like cells, representing one of three newly identified FCER1G⁺ innate-like CD8A⁺ T cell populations in the peripheral blood of healthy dogs. In conclusion, using scRNAseq combined with immune repertoire sequencing we were able to resolve canine TCRαβ⁺ T cell populations at unprecedented resolution. The peripheral blood TCRαβ⁺ T cell atlas of healthy dogs generated here represents an important reference data set for future studies and is of relevance for identifying new targets for T cell-specific therapies.