The effects of training on performance have focused mainly on the aerobic rather than the anaerobic component. PURPOSE: To compare the effects of nine weeks of anaerobic cycle ergometer training on absolute (PP) and relative (PP/KG) peak power of young male adults. METHODS: Thirty-nine college students were randomly assigned to one of the three experimental (EG) and one non-exercising control group (CG - n = 10; age = 22.4 ± 3.4 year). The EGs trained 3 times per week for 9 weeks, at maximum, with a workload of 7.5% of their respective body weights. The total relative work was equated among groups: EG1 −3 sprints of 30 s with 3 min of active recovery between sprints (n = 10, age = 20.6 ± 2.5 years); EG2 −3 sprints of 30 s with 5 min of active recovery between sprints (n = 10, age = 22.1 ± 3.4 years); EG3 −9 sprints of 10 s with a 3 min active recovery between sprints (n = 9; age = 21.0 ± 3.5 years). RESULTS: A factorial ANOVA with repeated measures (4×2) followed by Tukey is test for unequal samples showed that all EGs increased significantly (p ≤0.05) in PP (EG1: 657.6 ± 119.4 and 826.5 ± 109.8; EG2: 675.8 ± 112.3 and 789.1 ± 112.7; EG3: 688.5 ± 94.0 and 797.5 ± 128.0) and PP/KG (EG1: 10.2 ± 1.3 and 12.8 ± 1.1; EG2: 10.4 ± 1.1 and 12.2 ± 1.0; EG3: 10.6 ± 0.9 and 12.0 ± 1.0) from the pre-training values. No significant differences were observed among the EGs at post-training. CONCLUSIONS: The present results suggest that anaerobic cycle training with workloads of 7.5% of body weight, for nine weeks, may elicit significant gains in PP and PP/KG. Nevertheless, when the total relative work time is maintained constant, training with diferent stimuli and intervals, may not elicit better improvements in anaerobic power.