AtlasFold-260703

AtlasFold-260703 is a single-sequence protein structure prediction model built on AtlasLM-3B. It predicts monomer structures without requiring a multiple sequence alignment (MSA).

Installation

pip install "atlasfold[fold]"

Optional cuEquivariance kernels can be installed with:

pip install "atlasfold[fold,cuequiv]"

Usage

The AtlasFold and AtlasLM weights are downloaded automatically from Hugging Face:

from atlasfold.pretrained import load_model
from atlasfold.runner import FoldingRunner

model = load_model("atlasfold-260703", device="cuda")
runner = FoldingRunner(model)
result = runner.fold("example", "MKTAYIAKQRQISFVKSHFSRQDILDLWIYHTQGYFPD")
print(result.best.avg_plddt)

Command-line inference also downloads the weights automatically:

python run_atlasfold.py \
    --model monomer \
    --input-fasta monomers.fasta \
    --out-dir predictions/monomers

Use --cache-dir PATH to select a cache location. --model-path PATH remains available as an optional local checkpoint override.

See the AtlasFold repository for complete CLI and Python API documentation.

Files

• weights/atlasfold-260703.pth: AtlasFold monomer state dict.
• SHA256: cbadb227d40e801a3d268d6e884e6ed3c6b9fb89d558739fcf5970492732e5ea

License

The source code and model weights are released under the MIT License.

Citation

If you use this model in your research, please cite:

@article{seo2026atlasfold,
  author = {Seo, Seonghwan and Kim, Hyeongwoo and Moon, Seokhyun and Kim, Woo Youn and {Team KAIST}},
  title = {AtlasFold: Protein structure prediction with metagenomic-scale language models},
  year = {2026},
  doi = {10.64898/2026.09.04.749352},
  URL = {https://www.biorxiv.org/content/10.64898/2026.09.04.749352v2},
  journal = {bioRxiv}
}