Objective: To compare peak oxygen uptake (VO2peak) and the test-retest reliability of physiological parameters between a 1-min and a 3-min closed-end and an incremental open-end upper-body poling test.Methods: On two separate test days, 24 healthy, upper-body trained men (age: 28.3 ± 9.3 years, body mass: 77.4 ± 8.9 kg, height: 182 ± 7 cm) performed a 1-min, a 3-min and an incremental test to volitional exhaustion in the same random order. Respiratory parameters, heart rate (HR), blood lactate concentration (BLa), rating of perceived exertion (RPE), and power output were measured. VO2peak was determined as the single highest 30-s average. Relative reliability was assessed with the intra-class correlation coefficient (ICC2, 1) and absolute reliability with the standard error of measurement (SEM) and smallest detectable change (SDC).Results: The incremental (3.50 ± 0.46 L·min−1 and 45.4 ± 5.5 mL·kg−1·min−1) and the 3-min test (3.42 ± 0.47 L·min−1 and 44.5 ± 5.5 mL·kg−1·min−1) resulted in significantly higher absolute and body-mass normalized VO2peak compared to the 1-min test (3.13 ± 0.40 L·min−1 and 40.4 ± 5.0 mL·kg−1·min−1) (all comparisons, p < 0.001). Furthermore, the incremental test resulted in a significantly higher VO2peak as compared to the 3-min test (p < 0.001). VO2peak was significantly higher on day 1 than day 2 for the 1-min test (p < 0.05) and displayed a trend toward higher values on day 2 for the incremental test (p = 0.07). High and very high ICCs across all physiological parameters were found for the 1-min (0.827–0.956), the 3-min (0.916–0.949), and the incremental test (0.728–0.956). The SDC was consistently small for HR (1-min: 4%, 3-min: 4%, incremental: 3%), moderate for absolute and body-mass normalized VO2peak (1-min: 5%, 3-min: 6%, incremental: 7%) and large for BLa (1-min: 20%, 3-min: 12%, incremental: 22%).Conclusions: Whereas both the 3-min and the incremental test display high relative reliability, the incremental test induces slightly higher VO2peak. However, the 3-min test seems to be more stable with respect to day-to-day differences in VO2peak. The 1-min test would provide a reliable alternative when short test-duration is desirable, but is not recommended for testing VO2peak due to the clearly lower values.