The objective of this study was to test the accuracy of a milk leukocyte differential tester (QScout, AAD Inc., Morrisville, NC). The study was conducted at the University of Kentucky Coldstream Dairy. Milk samples from individual quarters (n = 320) were aseptically collected into clear, 150-mL polypropylene vials. Each sample was divided into 2 replicas after mixing. One replica was analyzed for SCC immediately following milking with a SomaCount FC (Bentley Instruments, Inc., Chaska, MN) using flow cytometry. Another replica sample was evaluated with the on-farm milk leukocyte differential device directly after milking, where total leukocyte count were obtained. Somatic cell count values and leukocyte counts were used to calculate SCS (SCS = log₂ (SCC/100) + 3). Subclinical mastitis thresholds of >200,000 (low) and >400,000 (high) cells/mL were set. Total cells measurements and calculated SCS from the milk leukocyte differential device were compared with SomaCount FC using correlation and regression coefficient of determination analyses. Also, sensitivity (Se), specificity (Sp), and accuracy were calculated for correct diagnosis of the 2 SCC thresholds using the SomaCount FC as reference. A correlation coefficient (r) of 0.97 (P < 0.001) was found for total leukocyte count, and r of 0.90 (P < 0.001) for SCS. In the regression the coefficient of determination was very high for SCC measures (R² = 0.94, MSE: 0.007; P < 0.001) and high for SCS (R² = 0.80, MSE:0.73; P < 0.001). The sensitivity (percentage of milk samples classified correctly) of the on-farm test was 58 [95% CI: 44 to 71]% and 72 [56 to 86]% for the low and high thresholds, respectively. The specificity (percentage of negative samples classified correctly) for the on-farm test was 100 [99 to 100]% and 100 [98 to 100]%, for the low and high thresholds, respectively. The overall accuracy (percentage of samples diagnosed correctly) was 93 [89 to 95]% and 96 [92 to 98]%, for the low and high thresholds, respectively. On-farm total leukocyte count is a promising technology that could support on farm mastitis management decisions.