<p>Over the last years, we have seen growing concerns on the need to publish computer code as an integral part of the research process. This has been reflected on improved publishing policies by scientific journals, addressing the relevant issues such as repositories or licensing. Here we explore the state-of-the-art of code availability and sharing for climate models, using as testbed the models from the Climate Model Intercomparison Project 5 and make some reflections on it. Our results show that there are great limitations in the access to the code of these climate models and that the climate modelling community needs to greatly improve their code sharing practices in order to comply with the best scientific practices and the most up to date editorial publishing policies.</p>