AlphaFold
GoogleAlphaFold is Google DeepMind's biomolecular structure prediction line. AlphaFold 3, the current tier, predicts the joint structure of complexes of proteins, nucleic acids, ligands, ions and modified residues from sequence together with genetic-search and template inputs. The inference source is Apache-2.0 in both generations, but the model parameters are licensed separately, and separately in each generation.
The slug covers the product line at the tier DeepMind ships, which is AlphaFold 3 in google-deepmind/alphafold3; the declared repository is the AlphaFold 2 codebase the line began with, and it is still the more widely starred of the two.
Openness
2 high confidence- weights
- open(AlphaFold 3 parameters at a published URL but only if received directly from Google
- data
- closed(training corpus not released in either generation)
- code
- partial(Apache-2.0 inference pipeline in both repositories
- license
- AlphaFold-3-Model-Parameters-Terms-of-Use(non-commercial use only
Google DeepMind distributes AlphaFold 3 parameters only on direct request, under terms that forbid commercial use, while AlphaFold 2's parameters remain freely available under CC-BY-4.0. AlphaFold 3 is the version now shipped, so its more restrictive terms are what the license reflects, keeping the line at restricted rather than open weights.
- https://raw.githubusercontent.com/google-deepmind/alphafold/main/README.md recorded 2026-10-06
"The AlphaFold parameters are made available under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) license"; the repository publishes inference only and points training data detail at the paper.
- https://raw.githubusercontent.com/google-deepmind/alphafold3/main/README.md recorded 2026-10-06
"You may only use AlphaFold 3 model parameters if received directly from Google"; parameters download from storage.googleapis.com/alphafold3/af3.bin.zst subject to separate terms; source code is Apache-2.0; the package "provides an implementation of the inference pipeline".
- https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md recorded 2026-10-06
The grant covers "non-commercial use by, or on behalf of, non-commercial organizations", provided "you are not a commercial organisation or acting on behalf of a commercial organisation", and excludes commercial activities including research on behalf of commercial organizations.
Adoption
3 medium confidenceNeither AlphaFold generation publishes a download count, so GitHub stars are the fallback, split between the original repository and the AlphaFold 3 repository that now holds the shipping parameters. A star marks interest, not use, so the count only approximates adoption.
- https://api.github.com/repos/google-deepmind/alphafold recorded 2026-09-10
stargazers_count: 14844; archived: false; fork: false.
- https://api.github.com/repos/google-deepmind/alphafold3 recorded 2026-09-10
stargazers_count: 8538; archived: false; fork: false.
Capability
2 high confidenceOne readable attribute. The breadth people associate with AlphaFold 3 is breadth of entity types in a single prediction task, and the category's note is explicit that a long task list is not an adaptation pathway - the repository ships inference only, and the model predicts a fixed target rather than designing anything.
- https://raw.githubusercontent.com/google-deepmind/alphafold3/main/README.md recorded 2026-09-10
Data pipeline runs genetic and template search alongside the input sequences; the package implements the inference pipeline only, with no training or fine-tuning entry point.
Verified 2026-09-10