Description
GT4SD (https://github.com/GT4SD/gt4sd-core) is an open-source library to accelerate hypothesis generation in the scientific discovery process that eases the adoption of state-of-the-art generative AI. In this session, we will present GT4SD code base and its main functionalities, ranging from model inference to training and using live applications demos in the context of material science and chemistry. The session is designed to provide a deep dive for developers and research scientists who want to accelerate their discovery pipelines with generative modeling capabilities.