After several years of extensive development, we now have a working website (https://openkim.org) with a repository and a processing pipeline that automatically couples interatomic models (potentials and force fields) and tests (simulations that compute material properties) to generate predictions. We also have an alpha version of a visualization framework to view model predictions and to compare them with other models and reference data. Under the hood, we have developed a comprehensive and extendible data structure for rigorously defining material properties. The KIM repository currently has 154 interatomic models and 1698 tests. Users can add new models, tests and visualizers.
To learn more about the KIM project and explore the features described above visit our "Getting Started with KIM" guide at https://openkim.org/doc/overview/getting-started/
This is a very exciting time for the KIM project. The launch of our new website, with the key core functionality of automatic model-test computations, is a major milestone. This is also the first step in extending model and test development to the wider research community. Our plan for the coming year is to get as many people involved as possible to make the project self-sustaining. We hope that everyone will join us in this emerging molecular simulation cyberinfrastructure.