Abstract We have developed and validated a new method to measure simultaneously glucose turnover, alanine turnover, and gluconeogenesis in human, in steady and non-steady states, using a double stable-isotope-labeled tracer infusion and GC-MS analysis. The method is based on the concomitant infusion and dilution of d-[2,3,4,6,6- 2H 5]glucose and l-[1,2,3- 13C 3]alanine. The choice of the tracers was done on the basis of a minimal overlap between the ions of interest and those arising from natural isotopic abundances. Alanine was chosen as the gluconeogenic substrate because it is the major gluconeogenic amino acid extracted by the liver and, with lactate, constitutes the bulk of the gluconeogenic precursors. The method was validated by comparing the results obtained during simultaneous infusion of trace amounts of both stable isotope labeled compounds with the radioactive tracers ( d-[3- 3H]glucose and l-[1,2,3- 14C 3]alanine) in a normal and a diabetic subject; the radiolabeled tracers were used as the accepted reference procedure. A slight overestimation of glucose turnover (7.3 versus 6.8 in normal and 10.8 versus 9.2 μmol/kg min in diabetic subject) was noticed when the stable isotope-labeled tracers were used. For the basal turnover rate of alanine, similar values were obtained with both methods (6.2 μmol/kg min). For gluconeogenesis, higher values were observed in the basal state with the stable isotopes (0.42 versus 0.21 μmol/kg min); however, these differences disappeared in the postprandial period after the ingestion of a mixed meal. Despite those minor differences, the overall correlation with the reference method was excellent for glucose turnover ( r = 0.87) and gluconeogenesis ( r = 0.86). These results indicate that the double stable isotope-labeled tracer technique is a reliable, safe, and acceptable method to evaluate those three metabolic processes in human in a single experiment.