<jats:p>Bovine milk small extracellular vesicles (sEVs) contain many biologically important molecules, including mRNAs. Quantitative real-time polymerase chain reaction (qRT-PCR) is a widely used method for quantifying mRNA in tissues and cells. However, the use, selection, and stability of suitable putative internal control genes in bovine milk sEVs for normalization in qRT-PCR have not yet been identified. Thus, the aim of the present study was to determine suitable putative internal control genes in milk sEVs for the normalization of qRT-PCR data. Milk sEVs were isolated from six healthy Holstein-Friesian cattle, followed by RNA extraction and cDNA synthesis. In total, 17 mRNAs were selected for investigation and quantification using qRT-PCR; they were further evaluated using geNorm, NormFinder, BestKeeper, and ∆CT algorithms to identify those that were highly stable putative internal control genes in milk sEVs. The final ranking of suitable putative internal control genes was determined using RefFinder. The mRNAs from TUB, ACTB, DGKZ, ETFDH, YWHAZ, STATH, DCAF11, and EGFLAM were detected in milk sEVs from six cattle by qRT-PCR. RefFinder demonstrated that TUB, ETFDH, and ACTB were highly stable in milk sEVs, and thus suitable for normalization of qRT-PCR data. The present study suggests that the use of these genes as putative internal control genes may further enhance the robustness of qRT-PCR in bovine milk sEVs. Since these putative internal control genes apply to healthy bovines, it would be helpful to include that the genes were stable in sEVs under “normal or healthy conditions”.</jats:p>