Phyllosilicates may trap hydrogen (H2) in the crust, but they may also produce it through various processes, including oxidative dehydrogenation. The dehydrogenation temperature depends on the type and composition of the phyllosilicates considered, but it may be as low as 300 °C. Here, we document the release of H2 and CO during thermal treatment of chloritite (300 °C) and talc (500 °C) from the Trimouns deposits (Eastern Pyrenees, France). Thermal release of gases coupled to stable isotope analysis has been used to recover and characterize H2 and CO, the two detected gases. Hydrogen content may be as high as 7 ppm with δDH2 values ranging from −258‰ to – 224‰ for sub-pure chloritite and − 140‰ for pure talc. CO content ranges between 3 ppm and 35.3 ppm with very homogeneous δ13CCO values between −27.6‰ and − 25.7‰. This study supports the idea that H2 was produced during experiments by dehydrogenation. The origin of CO remains enigmatic, but its carbon isotope composition suggests a link to the few amounts of graphite documented in chloritite and talc from the deposit. This work also reports extensive hydrogen isotope fractionation between H2 produced by dehydrogenation and both talc and chloritite. Dehydrogenation of phyllosilicates is a potential source term of H2 in numerous magmatic-hydrothermal settings and must thus be accounted for in the budget of the H2 geochemical cycle