To explore the treatment outcome of the strategy of early extubation and then switching to non-invasive mechanical ventilation in children with acute respiratory failure, and the safety and feasibility of using the strategy to replace traditional methods. A total of 102 children, aged between 1 month to 14 years old, who had acute respiratory failure and were admitted to the pediatric ICU of West China Second University Hospital, Sichuan University between January 2019 and December 2020 were enrolled and randomly assigned to treatment group 1 (n=55) and treatment group 2 (n=47). In addition, 53 children who had the same condition in the 12 month prior to the beginning of the study were included in the control group. In the two treatment groups, the patients were extubated first, and then weaned off the ventilator. In group 1, when the patient met the invasive-non-invasive switching criteria, the tracheal tube was pulled out and non-invasive bi-level positive airway pressure (BiPAP) ventilation was used for respiratory support. In group 2, high-flow nasal cannula (HFNC) oxygen therapy was used for respiratory support. The traditional progressive weaning method was adopted for the control group (extubing and weaning were performed at the same time). The incidence of ventilator-associated pneumonia (VAP) during the period of tracheal intubation was compared and the mortality of the two groups was evaluated from the point when the patients were recruited. At the time of extubation in the treatment groups and extubation plus weaning in the control group, the pressure support levels, or PC above PEEP, intubation time, sequential time (between 2 treatment groups only), weaning failure rate, and the incidence of laryngeal edema and nasal pressure ulcer were compared. The subjects of the study were predominantly infants (93 cases, 60%) and young children (31 cases, 20%). Among the 155 cases, 82 (53%) were male. There was no statistical difference in age distribution or gender among the groups. There was no significant difference in the clinical indicators among the three groups before tracheal intubation. At the time of extubation, the PC above PEEP in the two treatment groups was higher than that in the control group, and higher in group 1 than that of group 2, the difference being statistically significant (P<0.05). The intubation time of the two treatment groups was shorter than that of the control group, and shorter in group 1 than that of group 2 (P<0.05). The sequential time of group 2 was shorter than that of group 1 (P<0.05). The extubation failure rate and the incidence of VAP in the two treatment groups were lower than those in the control group, and there was no statistically significant difference between the two treatment groups. The incidence of nasal pressure ulcers in group 1 was higher than that in the other two groups (P<0.05). There was 1 death in treatment group 1, and no deaths in treatment group 2 or the control group. There was no significant difference in mortality or the incidence of laryngeal edema after extubation in the three groups. Early extubation and then switching to non-invasive mechanical ventilation can be well tolerated by the patients, and can be used in clinical practice as an effective weaning method for children with acute respiratory failure.