The potential interaction between metformin and exercise on glucose-lowering effects remains controversial. We studied the separated and combined effects of metformin and/or exercise on fasting and postprandial insulin sensitivity in individuals with pre-diabetes and type 2 diabetes (T2D). Eight T2D adults (60 ± 4 yr) with overweight/obesity (32 ± 4 kg·m-2) under chronic metformin treatment (9 ± 6 yr; 1281 ± 524 mg·day-1) underwent four trials; 1) taking their habitual metformin treatment (MET), 2) substituting during 96 h their metformin medication by placebo (PLAC), 3) placebo combined with 50 min bout of high-intensity interval exercise (PLAC + EX), and 4) metformin combined with exercise (MET + EX). Plasma glucose kinetics using stable isotopes (6,6-2H2 and [U-13C] glucose), and glucose oxidation by indirect calorimetry, were assessed at rest, during exercise, and in a subsequent oral glucose tolerance test (OGTT). Postprandial glucose and insulin concentrations were analyzed as mean and incremental area under the curve (iAUC), and insulin sensitivity was calculated (i.e., MATSUDAindex and OGISindex). During OGTT, metformin reduced glucose iAUC (i.e., MET and MET + EX lower than PLAC and PLAC + EX, respectively; P = 0.023). MET + EX increased MATSUDAindex above PLAC (4.8 ± 1.4 vs. 3.3 ± 1.0, respectively; P = 0.018) and OGISindex above PLAC (358 ± 52 vs. 306 ± 46 mL·min-1·m-2, respectively; P = 0.006). Metformin decreased the plasma appearance of the ingested glucose (Ra OGTT; MET vs. PLAC, -3.5; 95% CI -0.1 to -6.8 µmol·kg-1·min-1; P = 0.043). Metformin combined with exercise potentiates insulin sensitivity during an OGTT in individuals with pre-diabetes and type 2 diabetes. Metformin's blood glucose-lowering effect seems mediated by decreased oral glucose entering the circulation (gut-liver effect) an effect partially blunted after exercise.NEW & NOTEWORTHY Metformin is the most prescribed oral antidiabetic medicine in the world but its mechanism of action and its interactions with exercise are not fully understood. Our stable isotope tracer data suggested that metformin reduces the rates of oral glucose entering the circulation (gut-liver effect). Exercise, in turn, tended to reduce postprandial insulin blood levels potentiating metformin improvements in insulin sensitivity. Thus, exercise potentiates metformin improvements in glycemic control and should be advised to metformin users.