HelixFold 3

Biological & Workflow Fit

Biological Application

Target-ligand interaction modeling, RNA-protein complex prediction, and covalent drug discovery

Research Workflow

Input biomolecular sequence and SMILES -> generate 3D complex coordinates and confidence scores

Compute & Hardware

1x NVIDIA A100 or Kunlunxin AI accelerators

Licensing & Academic Use

Apache 2.0 (code) / Model License

Documented Evidence

View validation publication / source ↗

Use this citation format when referencing HelixFold 3 in scientific publications and benchmark papers.

@software{helixfold_3_2026,
  title = {{HelixFold 3}},
  author = {{Baidu PaddleHelix}},
  year = {2026},
  url = {https://github.com/PaddlePaddle/PaddleHelix},
  note = {Indexed on aibioatlas - AI for Biology and Drug Discovery}
}