Purpose: This study investigated the effects of repeated-sprint (RS) training in hypoxia induced by voluntary hypoventilation at low lung volume (RSH-VHL) on physiological adaptations, RS ability (RSA) and anaerobic performance. Methods: Over a 3-week period, eighteen well-trained cyclists completed six RS sessions in cycling either with RSH-VHL or with normal conditions (RSN). Before (Pre) and after (Post) the training period, the subjects performed an RSA test (10 × 6-s all-out cycling sprints) during which oxygen uptake (V ˙ O 2) and the change in both muscle deoxyhaemoglobin ($Δ$[HHb]) and total haemoglobin ($Δ$[THb]) were measured. A 30-s Wingate test was also performed and maximal blood lactate concentration ([La]max) was assessed. Results: At Post compared to Pre, the mean power output during both the RSA and the Wingate tests was improved in RSH-VHL (846 ± 98 vs 911 ± 117 W and 723 ± 112 vs 768 ± 123 W, p < 0.05) but not in RSN (834 ± 52 vs 852 ± 69 W, p = 0.2; 710 ± 63 vs 713 ± 72 W, p = 0.68). The average V ˙ O 2 recorded during the RSA test was significantly higher in RSH-VHL at Post but did not change in RSN. No change occurred for $Δ$[THb] whereas $Δ$[HHb] increased to the same extent in both groups. [Lamax] after the Wingate test was higher in RSH-VHL at Post (13.9 ± 2.8 vs 16.1 ± 3.2 mmol L−1, p < 0.01) and tended to decrease in RSN (p = 0.1). Conclusions: This study showed that RSH-VHL could bring benefits to both RSA and anaerobic performance through increases in oxygen delivery and glycolytic contribution. On the other hand, no additional effect was observed for the indices of muscle blood volume and O2 extraction.