Objective: evidence is inconsistent for the effect of non-invasive ventilation (NIV) for individuals with chronic obstructive pulmonary disease (COPD) during exercise training. This review aimed to determine the effect of NIV in COPD individuals during exercise training on exercise capacity, quality of life, functional performance and symptoms. Material and methods: we searched for studies evaluating the effect of NIV on COPD individuals during exercise training published until May 2020 in 6 electronic databases (PubMed, Embase, Cochrane Library, Web of Science, clinical trial registers and Wanfang). The included studies were appraised with the Cochrane Risk of Bias tool and Downs and Black criteria. The primary outcomes were improvement in 6-min walking distance and quality of life. Mean difference (MD) or standardized mean difference (SMD) with 95% confidence intervals (CIs) was calculated. Results: among 855 identified articles, reports for 15 studies with heterogeneous populations were eligible, with 520 individuals: 257 in the NIV group and 263 in the control group. Across studies, NIV intervention during exercise training affected exercise performance (6-min walking distance: SMD: 0.33, 95% CI: 0.06; 0.59, P=0.02; quality of life: SMD: −0.77, 95% CI: −1.01; −0.53, P<0.001). In the analysis of dyspnea, pooled estimates demonstrated improvement in the NIV versus control group. NIV intervention was also better than exercise alone in ameliorating oxygen saturation, PaO2, PaCO2, blood lactate level and breath rate. The groups did not differ in duration of exercise, BODE index, minute ventilation, heart rate and systolic blood pressure. Conclusions: our review suggests that NIV is a relevant adjuvant for exercise training in COPD individuals because the intervention could improve exercise performance and quality of life. The current results also demonstrate the importance of further investigations of higher methodological quality to assess the effect on exercise capacity and quality of life.