Purpose: Interval-training sets may be applied in a different sequence within a swimming training session. The aim of this study was to investigate the effect of different set sequences on performance and physiological responses in a training session. Methods: Twelve highly trained male swimmers performed 4 sessions in randomized order. Each session included a different combination of 2 training sets: set A–set C, set C–set A, set B–set C, or set C–set B. Set A consisted of 8 × 200 m at speed corresponding to lactate threshold (30-s recovery), set B included 8 × 100 m at maximum aerobic speed (30-s recovery), and set C included 4 × 50-m all-out swimming (2-min recovery). Performance and physiological responses (lactate concentration, pH, base excess, bicarbonate, heart rate, and heart-rate variability) were measured. Results: Performance in each set was similar between sessions irrespective of set sequence. Blood lactate, heart rate, and acid–base responses during set C were similar in all sessions, but blood lactate was higher in sets A and B during C–A and C–B sessions (P = .01). The overall blood lactate and acid–base response was higher in C–A and C–B sessions compared with A–C and B–C sessions, respectively (P = .01). Heart-rate variability in each set, separately as well as the overall session effect, did not differ and was thus independent to the set sequence applied. Conclusions: Training sessions including all-out swimming as a first set increase the magnitude of metabolic responses to the subsequent aerobic-dominated training set.