RFdiffusion
Rosetta CommonsRFdiffusion is a diffusion model for de novo protein design from the Institute for Protein Design, built on RoseTTAFold's architecture and trained parameters. It generates unconditional monomers, scaffolds functional motifs, designs binders against a target with specified hotspot residues, builds symmetric oligomers, and partially rediffuses an existing backbone to diversify it.
Openness
3 high confidence- weights
- open(checkpoints downloaded from files.ipd.uw.edu and covered explicitly by the repository LICENSE)
- data
- closed(training set not released
- code
- partial(inference scripts and hydra configs only
- license
- BSD-3-Clause(OSI)
GitHub reports NOASSERTION, which would have left this product looking unlicensed. The LICENSE body is BSD three-clause text under a University of Washington copyright, and it says outright that it covers the model weights as well as the source, which is the rarer and more useful of the two statements.
- https://api.github.com/repos/RosettaCommons/RFdiffusion recorded 2026-09-10
license spdx_id: NOASSERTION, key "other", url null - the endpoint does not recognise the BSD text in the LICENSE file.
- https://raw.githubusercontent.com/RosettaCommons/RFdiffusion/main/LICENSE recorded 2026-09-10
"BSD License. Copyright (c) 2023 University of Washington ... This copyright and license covers both the source code and model weights referenced for download in the README file", followed by the three standard BSD conditions.
- https://raw.githubusercontent.com/RosettaCommons/RFdiffusion/main/README.md recorded 2026-09-10
"RFdiffusion is released under an open source BSD License (see LICENSE file). It is free for both non-profit and for-profit use." Weights are fetched with wget from files.ipd.uw.edu; the published code covers inference configuration only, and no training data or pipeline is released.
Adoption
2 medium confidenceWeights are served from an institutional file host rather than the Hub, so there is no download count; GitHub stars are the fallback signal instead.
- https://api.github.com/repos/RosettaCommons/RFdiffusion recorded 2026-09-10
stargazers_count: 3043; archived: false; fork: false.
Capability
2 high confidenceRFdiffusion documents one readable capability that most of this category lacks: it designs new protein backbones rather than predicting a structure for an existing sequence. Its conditioning inputs, motifs, hotspot residues, symmetry, an existing structure, are varied but all derive from the same PDB structure format, so they count as one input system rather than several.
- https://raw.githubusercontent.com/RosettaCommons/RFdiffusion/main/README.md recorded 2026-09-10
Documents unconditional monomer generation, motif scaffolding, binder design against a target with hotspot residues, symmetric oligomer design and partial diffusion of an existing backbone; conditioning comes from an input .pdb with contigs, secondary structure and block adjacency, and the repository publishes inference only.
Verified 2026-09-10