ABSTRACT: The main aims of this study were to determine the accuracy of the portable metabolic cart K5 by comparison with a stationary metabolic cart (Vyntus CPX), to check on the validity of Vyntus CPX using a butane combustion test, and to assess the reliability of K5 during prolonged walks in the field. For validation, measurements were consecutively performed tests with both devices at rest and during submaximal exercise (bicycling) at low (60 W) and moderate intensities (130-160 W) in 16 volunteers. For the reliability study, 14 subjects were measured two times during prolonged walks (13 km, at 5 km/h), with the K5 set in mixing chamber (Mix) mode. Vyntus measured the stoichiometric RQ of butane combustion with high accuracy (error <1.6%) and precision (CV <0.5%), at VO2 values between 0.788 and 6.395 L/min. At rest and 60 W, there was good agreement between Vyntus and K5 (breath-by-breath, B×B) in VO2, VCO2, RER, and energy expenditure, while in Mix mode the K5 overestimated VO2 by 13.4 and 5.8%, respectively. Compared to Vyntus, at moderate intensity the K5 in B×B mode underestimated VO2, VCO2, and energy expenditure by 6.6, 6.9, and 6.6%, respectively. However, at this intensity there was an excellent agreement between methods in RER and fat oxidation. In Mix mode, K5 overestimated VO2 by 5.8 and 4.8%, at 60 W and the higher intensity, respectively. The K5 had excellent reliability during the field tests. Total energy expenditure per Km was determined with a CV for repeated measurements of 4.5% (CI: 3.2-6.9%) and a concordance correlation coefficient of 0.91, similar to the variability in VO2. This high reproducibility was explained by the low variation of FEO2 measurements, which had a CV of 0.9% (CI: 0.7-1.5%) combined with a slightly greater variability of FECO2, VE, VCO2, and RER. In conclusion, the K5 is an excellent portable metabolic cart which is almost as accurate as a state-of-art stationary metabolic cart, capable of measuring precisely energy expenditure in the field, showing a reliable performance during more than 2 h of continuous work. At high intensities, the mixing-chamber mode is more accurate than the B×B mode.