Abstract Phosphorous (31 P) magnetization transfer (MT) experiments enable the non-invasive investigation of human muscle metabolism in various physiological and pathological conditions. The purpose of our study was to investigate the feasibility of time-resolved MT, and to compare the results of MT experiments at 3 T and 7 T. Six healthy volunteers were examined on a 3 T and a 7 T MR scanner using the same setup and identical measurement protocols. In the calf muscle of all volunteers, four separate MT experiments (each ∼10 min duration) were performed in one session. The forward rate constant of the ATP synthesis reaction ( kATP ) and creatine kinase reaction ( kCK ), as well as corresponding metabolic fluxes ( FATP , FCK ), were estimated. A comparison of these exchange parameters, apparent T1 s, data quality, quantification precision, and reproducibility was performed. The data quality and reproducibility of the same MT experiments at 7 T was significantly higher (i.e., kATP 2.7 times higher and kCK 3.4 times higher) than at 3 T ( p < 0.05). The values for kATP ( p = 0.35) and kCK ( p = 0.09) at both field strengths were indistinguishable. Even a single MT experiment at 7 T provided better data quality than did a 4 times-longer MT experiment at 3 T. The minimal time-resolution to reliably quantify both FATP and FCK at 7 T was ∼6 min. Our results show that MT experiments at 7 T can be at least 4 times faster than 3 T MT experiments and still provide significantly better quantification. This enables time-resolved MT experiments for the observation of slow metabolic changes in the human calf muscle at 7 T.