Oxygen gas flowmeters (OGF) are used to regulate the oxygen flow in acute and chronic care. In hospitals, Thorpe tubes (TT) are the classical systems most used for delivering oxygen. In recent years, the oxygen flow restrictor (OFR) has appeared. These devices use a series of calibrated openings in a disk that can be adjusted to deliver different flow rates. These devices have a reputation for delivering more accurate oxygen flow rates compared to classical OGFs. However, to our knowledge, few study has examined this supposition. This study aimed to compare and evaluate the accuracy and precision of the ready-to-use TTs and OFRs. OGFs were selected from hospitals in Belgium and France. Before performing the flow measurements, the inlet pressure was checked. The accuracy of the OGF was analyzed with a calibrated thermal mass flowmeter (RED Y COMPACT™ GCM-0 to 20L/min-VÖGTLIN Instruments). Different flows (2, 4, 6, 9 or 12L/min) were evaluated. Linear regression analysis, bias (with confidence interval) and lower and upper limit of the agreement were calculated for TTs and OFRs. All measurements are expressed in absolute values. Four-hundred-seventy-six TTs and 96 OFRs were analyzed. The intra-class correlation coefficient calculated for the calibrated thermal mass flowmeter was > 0.99 and reflected the excellent reliability of our measurements. For TTs, the bias value was -0.24 L/min (± 0.88), and the limits of agreement were -1.97 to 1.48L/min. For OFRs, the bias value was -0.30 L/min (± 0.54), and the limits of agreement were -1.36 to 0.77L/min. As the flow increased, the accuracy of all analyzed OGFs decreased. With the increasing flow, some data fell outside the limits of agreement, and the trend increased with the elevated oxygen flow. TTs were less accurate compared to OFRs due to the increased flow variability. However, for TTs and OFRs, as the required flow is elevated, the dispersion of values increases on both sides of the actual flow.