The 3D-ABC project is an ambitious initiative launched by the Helmholtz Foundation Model Initiative (HFMI) to create foundation models aimed at mitigating global climate change through comprehensive monitoring of Earth's carbon cycle. This project brings together leading institutions within the Helmholtz Association, including HZDR, AWI, UFZ, FZJ, DLR, and GFZ. Led by Pedram Ghamisi, the AI team is developing these models using diverse data sources, such as satellite, airborne, drone, and ground-based sensors. The models are trained using self-supervised learning, allowing them to leverage large volumes of unlabeled data. They will then be fine-tuned for specific carbon stock assessment tasks, including land cover analysis, vegetation height estimation, and above-ground biomass measurement, among others.