Background and objective: Non-invasive ventilation (NIV) is part of the standard of care for hypercapnic respiratory failure secondary to COPD, but may be poorly tolerated. Preliminary evidence suggests nasal high-flow (NHF) therapy may improve hypercapnia in COPD and be well tolerated. We compared NHF and NIV in people with COPD and chronic hypercapnic respiratory failure. Methods: Single-blind randomized controlled two-way cross-over single-centre trial was conducted in New Zealand. Twenty-four participants with stable hypercapnic COPD received: NHF at 45 L/min and NIV at 15/4 cm H 2 O, each for 60 min with a 15-min washout in between. The primary outcome was transcutaneous partial pressure of carbon dioxide (PtCO 2 ) at 60 min, adjusted for baseline. Results: NIV reduced the PtCO 2 more than NHF (mean (SD) at 60 min by −5.3 (5.0) vs −2.5 (3.5) mm Hg; difference: −2.8 (−5.0 to −0.5) P = 0.021). Difference across all time points was −2.5 mm Hg (95% CI −4.5 to −0.5, P = 0.016). There was no significant difference in the proportion of participants with a reduction of PtCO 2 ≥ 4 or ≥ 8 mm Hg. Participants rated NHF significantly better for ease of application, comfort and fit. Conclusion: In stable COPD patients with chronic hypercapnia, NIV resulted in a greater reduction in PtCO 2 compared with NHF, which was of uncertain clinical significance. NHF was better tolerated than NIV and may be a therapeutic option for some people with hypercapnic respiratory failure. Clinical Trial Registration: ACTRN12616001701415 at www.anzctr.org.au