Background and Purpose: This meta-analysis quantified mean effects of moderate-to-vigorous intensity locomotor training (LTmv) on walking outcomes in subacute and chronic stroke, and the magnitude of individual variability in LTmv response. Methods: Databases were searched for randomized trials comparing LTmv with no intervention, non-gait intervention or low-intensity gait training. Comfortable gait speed (CGS), fastest gait speed (FGS), 6-minute walk distance (6MWT), walking activity (steps/day) and adverse effect/event (AE) data were extracted. Pooled estimates were calculated for mean changes within and between groups, the relative risk of different AEs, and the standard deviation of individual response (SDIR) to LTmv versus control groups, stratified by study chronicity where possible. Results: There were 19 eligible studies (total N=1,096); 14 in chronic stroke (N=839) and 5 in subacute stroke (N=257). Compared with control interventions, LTmv yielded significantly greater increases in CGS, FGS and 6MWT in both subacute and chronic stroke, with subacute studies showing significantly greater effect sizes for CGS, FGS and nearly 6MWT (p=0.054). In 4 studies reporting steps/day data, LTmv was not significantly different from control interventions. In 14 studies reporting on AEs, there were no treatment-related serious AEs among 398 LTmv participants. SDIR estimates indicated significant individual response variability for CGS, nearly FGS (p=0.0501) and 6MWT. Discussion and Conclusions: LTmv improves mean walking capacity outcomes in subacute and chronic stroke and does not appear to have high risk of serious harm, but response magnitude varies between chronicity subgroups and individuals, and few studies have tested effects on daily walking activity or non-serious AEs.