A single bout of intense aerobic exercise performed after the acquisition of a new sequence of movements has been shown to enhance retention. However, no such exercise-related advantage was reported when participants learned to adapt their reaching movements to compensate for a 60° deviation of the visual feedback (rVMA task). Since the sequence tasks used previously relied mainly on implicit learning processes, as opposed to the rVMA task which had an explicit component given the size of the deviation, it is possible that exercise-related gains in retention may be restricted to implicit learning. To test this hypothesis, we used a modified version of the rVMA task in which a smaller deviation (25°) was introduced gradually (1° per 10 trials), making it difficult to be perceived explicitly and thus favoring an implicit adaptation. Thirty participants took part in the experiment (7 males, 23 females; mean age: 22 ± 2.36) and, immediately after the rVMA acquisition session (500 trials in total), either performed 3 bouts of 3 minutes of cycling at 80% of their maximal aerobic power (Exercise Group) or watched a relaxation video (Control Group). When retention of the rVMA task was assessed 24 hours later in a no-vision test, participants of the Exercise group demonstrated better retention compared to those of the Control group (p = 0.037, d = 0.8). This result demonstrates that visuomotor adaptation can benefit from exercise and suggests that exercise-related effects on motor skill learning may be restricted to implicit learning.