High-flow nasal cannula (HFNC) and helmet noninvasive ventilation (NIV) are used for the management of acute hypoxemic respiratory failure. Physiological comparison of HFNC and helmet NIV in patients with hypoxemia. Fifteen patients with hypoxemia with Pa /Fi  < 200 mm Hg received helmet NIV (positive end-expiratory pressure ≥ 10 cm H O, pressure support = 10-15 cm H O) and HFNC (50 L/min) in randomized crossover order. Arterial blood gases, dyspnea, and comfort were recorded. Inspiratory effort was estimated by esophageal pressure (Pes) swings. Pes-simplified pressure-time product and transpulmonary pressure swings were measured. As compared with HFNC, helmet NIV increased Pa /Fi (median interquartile range: 255 mm Hg 140-299 vs. 138 101-172;  = 0.001) and lowered inspiratory effort (7 cm H O 4-11 vs. 15 8-19;  = 0.001) in all patients. Inspiratory effort reduction by NIV was linearly related to inspiratory effort during HFNC (  = 0.84;  < 0.001). Helmet NIV reduced respiratory rate (24 breaths/min 23-31 vs. 29 26-32;  = 0.027), Pes-simplified pressure-time product (93 cm H O ⋅ s ⋅ min 43-138 vs. 200 168-335;  = 0.001), and dyspnea (visual analog scale 3 2-5 vs. 8 6-9;  = 0.002), without affecting Pa (  = 0.80) and comfort (  = 0.50). In the overall cohort, transpulmonary pressure swings were not different between treatments (NIV = 18 cm H O 14-21 vs. HFNC = 15 8-19;  = 0.11), but patients exhibiting lower inspiratory effort on HFNC experienced increases in transpulmonary pressure swings with helmet NIV. Higher transpulmonary pressure swings during NIV were associated with subsequent need for intubation. As compared with HFNC in hypoxemic respiratory failure, helmet NIV improves oxygenation, reduces dyspnea, inspiratory effort, and simplified pressure-time product, with similar transpulmonary pressure swings, Pa , and comfort.