Persistent long-term memory depends on successful stabilization and integration of new memories after initial encoding [1, 2]. This consolidation process is thought to require neuromodulatory factors such as dopamine, noradrenaline, and brain-derived neurotrophic factor [3-7]. Without the release of such factors around the time of encoding, memories will decay rapidly [3, 5, 6, 8]. Recent studies have shown that physical exercise acutely stimulates the release of several consolidation-promoting factors in humans [9-14], raising the question of whether physical exercise can be used to improve memory retention [15-17]. Here, we used a single session of physical exercise after learning to exogenously boost memory consolidation and thus long-term memory. Three groups of randomly assigned participants first encoded a set of picture-location associations. Afterward, one group performed exercise immediately, one 4 hr later, and the third did not perform any exercise. Participants otherwise underwent exactly the same procedures to control for potential experimental confounds. Forty-eight hours later, participants returned for a cued-recall test in a magnetic resonance scanner. With this design, we could investigate the impact of acute exercise on memory consolidation and retrieval-related neural processing. We found that performing exercise 4 hr, but not immediately, after encoding improved the retention of picture-location associations compared to the no-exercise control group. Moreover, performing exercise after a delay was associated with increased hippocampal pattern similarity for correct responses during delayed retrieval. Our results suggest that appropriately timed physical exercise can improve long-term memory and highlight the potential of exercise as an intervention in educational and clinical settings.