Purpose This study examined the effect of 3 and 6 weeks of intensity domain-based exercise training on V˙O2 kinetics changes and their relationship with indices of performance. Methods Eighty-four young healthy participants (42 M, 42 F) were randomly assigned to six groups (14 participants each, age and sex-matched) consisting of: continuous cycling in the (1) moderate (MOD)-, (2) lower heavy (HVY1)-, and (3) upper heavy-intensity (HVY2)- domain; interval cycling in the (4) severe-intensity domain (i.e., high-intensity interval training (HIIT), or (5) extreme-intensity domain (i.e., sprint-interval training (SIT)); or (6) control (CON). Training participants completed two three-week phases of three supervised sessions per week, with physiological evaluations performed at PRE, MID and POST intervention. All training protocols, except SIT, were work-matched. Results There was a significant time effect for the time constant (τV˙O2) between PRE (31.6 ± 10.4 s) and MID (22.6 ± 6.9 s) (p textless 0.05) and PRE and POST (21.8 ± 6.3 s) (p textless 0.05), but no difference between MID and POST (p textgreater 0.05) and no group or interaction effects (p textgreater 0.05). There were no PRE to POST differences for CON (p textless 0.05) in any variables. Despite significant increases in maximal V˙O2 (V˙O2max), estimated lactate threshold (θLT), maximal metabolic steady state (MMSS), and peak power output (PPO) for the intervention groups (p textless 0.05), there were no significant correlations from PRE to MID or MID to POST between ΔτV˙O2 and ΔV˙O2max (r = – 0.221, r = 0.119), ΔPPO (r = – 0.112, r = – 0.017), ΔθLT (r = 0.083, r = 0.142) and ΔMMSS (r = – 0.213, r = 0.049)(p textgreater 0.05). Conclusion This study demonstrated that (i) the rapid speeding of V˙O2 kinetics was not intensity-dependent; and (ii) changes in indices of performance were not significantly correlated with ΔτV˙O2.