The effects of specific training protocols to improve repeated sprint ability are well known; however, the utility of non-specific training for this purpose is doubtful. Therefore, this study aimed to compare the effects of a specific (sport-related) vs. non-specific (general physical development) eight-week training protocol on developing or maintaining repeated sprint ability in soccer players. Thirty-eight male soccer players (aged 17.1 ± 0.3 years) were randomly divided into a speed endurance group (n = 18) and a core training group (n = 20). The speed endurance group performed 4 sprints lasting 15 seconds with a recovery of 75 seconds, while the core training group performed 2 sets of 4 exercises for 30 seconds with 15 seconds of recovery. Both protocols were performed twice a week. All players performed the Running-Based Anaerobic Sprint Test (6 consecutive 35-meter sprints with 10-seconds recovery intervals). A mixed two-way ANOVA was performed with no significant effects in the interaction term. On average, the speed endurance and the core training groups improved their repeated sprint ability by 7.9% and 8.0%, respectively, from pre- to post-assessment (p < 0.001, η²_p = 0.459). No significant between-group differences were detected for the Running-Based Anaerobic Sprint Test total time or maximum power. In principle, specific and non-specific low-volume training protocols may improve repeated sprint ability over eight weeks in soccer players. However, the improvements may have been derived from the continued soccer-specific training, which is the reason why future studies should include a passive control group and test whether high session volumes produce different results.

• Similar improvements in RSA were observed by adding a SET protocol and a core training protocol to regular soccer training, over eight weeks, in youth players.

• The improvements observed with the specific speed training were not superior to the non-specific core training, suggesting that non-specific and specific training may have a similar contribution to enhance RSA.

• In the absence of a passive control group, it is unclear to what extent the improvements in RSA observed were influenced by the regular soccer training applied to the whole team.