Microsoft Foundry

Experiment

RosettaFold3

RosettaFold3 (RF3) is a unified biomolecular modeling system that predicts 3D structures of proteins, nucleic acids, and small molecules within a single framework. Combining multimodal transformers and generative diffusion models, RF3 enables precise modeling of complex molecular assemblies such as protein–ligand, protein–DNA, and protein–RNA interactions. This flexibility supports rapid innovation in drug discovery, enzyme design, materials science, and beyond. RF3 was developed by the Baker lab and DiMaio lab from the Institute for Protein Design (IPD) at the University of Washington, in collaboration with Microsoft’s AI for Good lab and other research partners.

Try RosettaFold3 in Foundry Catalog RosettaFold3 has been released for research purposes.

Figure 1: Predicted structure of a protein in complex with DNA

Figure 2: Predicted structure of a protein in complex with 3 different chemical compounds

RF3 introduces atom-level conditioning and templating, allowing researchers to guide folding around specific ligands or impose experimental constraints directly into the model. This fine-grained control enables accurate structure prediction for custom-designed molecules and improves interpretability in structure-guided design tasks.

Built for extensibility, RF3 is the state-of-the-art open-source model comparable to leading models like AlphaFold3 (AF3) while providing greater flexibility, chirality handling, and support for mixed L/D peptides. Its unified architecture supports open customization for diverse molecular types and design objectives.

By reducing reliance on expensive experimental methods such as cryo-EM or X-ray crystallography, RF3 accelerates discovery cycles and lowers validation costs across research and industry. It allows scientists to focus resources on the most promising candidates, driving faster and more reliable R&D outcomes.

As a foundation model for multimolecular discovery, RF3 extends biomolecular AI beyond proteins—enabling enzyme design for sustainability, gene therapy optimization, and materials innovation for aerospace and biosecurity applications.