More than 140 molecular species have been observed in the interstellar medium since 1937 containing up to 13 atoms. In many cases, chemical models fail to reproduce observations. In cold dark clouds for instance, gas-phase methanol is observed while chemical models, even including gas-grain interactions, underestimated its abundance by several orders of magnitude. Although some of these discrepancies are probably due to the fact that these models do not take into account the history of the material or some important physical processes, many improvements can be done from the chemical point of view. Indeed, chemical models use networks in which only a small fraction of the reactions have been studied in interstellar medium conditions. Before making models more complicated, one has to get the chemistry right. For this reason, astrochemists really rely on experimental measurements and theoretical calculations of rate coefficients. Similarly, chemists interested in this field of research need astrochemists to guide them to the priority reactions to be studied. KIDA (for KInetic Database for Astrochemistry) was built with the aim of improving the exchange between these two communities: the astrophysicists and the chemists/physicists who determine the reaction rate coefficients. KIDA is a database, which will contain all the kinetic data that can be of interest for the chemical modeling of the interstellar medium and planetary atmospheres. In addition of containing information about the data (references, detailed on the methods to obtain the data, range of temperature etc), the interface with the database allows physicists and chemists to upload themselves their data. In order to provide help the non-chemists, a group of experts give an expertise on the quality of the data.< Réduire