The purpose of the present study was to investigate the effect of different durations of half-time re-warm up (RW) on intermittent sprint performance. Using a randomized crossover design, 13 healthy men performed three trials, which consisted of two, 40-min intermittent exercises separated by a 15-min half-time. Half-time interventions were 15 min of seated rest (Control), 7 min of cycling at 70% of maximal heart rate (HR_max) (7 min RW), and 3 min of cycling at 70% of HR_max (3 min RW). The second 40-min intermittent exercise as an exercise performance test was the Cycling Intermittent-Sprint Protocol (CISP), which consisted of 10 s of rest, 5 s of maximal sprint, and 105 s of low-intensity exercise at 50% of VO_2max, with the cycles repeated over the 40-min duration. The mean work during the maximal sprint in the initial 10 min of the CISP was higher in the both RW trials than in the control trial (control: 3638 ± 906 J, 7 min RW: 3808 ± 949 J, p < 0.05, 3 min RW: 3827 ± 960 J, p < 0.05). There were no significant differences among three trials for mean work at 10-20, 20-30, and 30-40 min of the CISP. In the initial 10 min of the CISP, the change in oxygenated hemoglobin concentration during the 105 s of exercise at 50% of VO_2max, oxygen uptake, carbon dioxide production, and respiratory exchange ratio were higher in both RW trials than in the control trial (p < 0.05). The rating of perceived exertion after half-time interventions was higher in both RW trials than in the control trial (p < 0.05). In conclusion, the 3 min RW increased intermittent sprint performance after the half-time, compared with a traditional passive half-time practice, and was as effective in improving intermittent sprint performance as the 7 min RW.