In horse breeding, quality assessment of semen before insemination is often requested. Non-laboratory-based techniques for objective analysis of sperm motility are thus of interest. The aim of this study was evaluating a portable device for semen analysis (Ongo sperm test) and its comparison with computer-assisted semen analysis (CASA). Semen was collected from 10 stallions, diluted to 100, 50 and 25 × 10  sperm/ml and analysed for total (TM) and progressive motility (PM). The final sperm concentration influenced total motility analysed by Ongo (p < 0.05) which was higher at 100 × 10  sperm/ml when compared to 25 × 10  sperm/ml (p < 0.05) but not when compared to 50 × 10  sperm/ml (n.s.). Sperm concentration did not influence total motility when assessed by SpermVision (n.s.). Agreement between methods was evaluated by correlation analysis and Bland-Altman plot. Intra-assay variation of Ongo was 5.2% ± 3.0 for TM and 6.9% ± 3.4 for PM. Correlation between Ongo and CASA was r = 0.79, 0.88 and 0.83 for 100, 50 and 25 × 10  sperm/ml for TM, and r = 0.87, 0.89 and 0.87 for PM, respectively (all p < 0.001). At the 100 and 25 mio/ml dilutions, the difference between the two systems deviated significantly from 0, while no such bias existed at the 50 mio/ml dilution (TM Ongo 85.0%, CASA 82.3%; PM Ongo 64.1%, CASA 66.1%). The 95% confidence interval was 19.9%, 18.9% and 19.2% ± mean for TM and 20.7%, 17.4% and 20.3% ± mean for 100, 50 and 25 × 10  sperm/ml, respectively. In conclusion, Ongo sperm test sperm motility data were strongly correlated with data obtained by CASA. In addition, at a concentration of 50 × 10  sperm/ml values measured with both systems were close to identical. At this concentration, which is recommended in equine AI, Ongo and CASA can be used interchangeably.