Protenix Base

protenix-base

OFoldX pipeline artifact for biomolecular structure prediction, using the alphafold3 architecture.

Disclaimer

This model card was generated by the OFoldX team for an OFoldX pipeline artifact. The upstream model authors did not write this card unless explicitly stated otherwise.

OFoldX is pre-alpha research software. Check the source checkpoint, upstream release, and local validation before using the artifact for scientific or operational decisions.

Model Details

AlphaFold 3-shaped structure-prediction model converted for the OFoldX runtime.

Converted Protenix v0.5.0 structure-prediction weights in an AlphaFold 3-shaped OFoldX implementation.

Model Provenance

• Upstream Project: Protenix
• Source Checkpoint: protenix_base_default_v0.5.0.pt
• Source Release: https://github.com/bytedance/Protenix/blob/main/docs/supported_models.md
• Primary Paper: Protenix - Advancing Structure Prediction Through a Comprehensive AlphaFold3 Reproduction
• Upstream License: Apache-2.0 for upstream Protenix code and model parameters

Model Specification

Field	Value
Repository	oteam/protenix-base
Artifact Kind	pipeline
Task	structure_prediction
Architecture	alphafold3
Entrypoint	ofoldx.pipelines.structure_prediction.StructurePredictionPipeline
Source Checkpoint	protenix_base_default_v0.5.0.pt

[!NOTE] Source checkpoint: protenix_base_default_v0.5.0.pt; v0.5.0 is the upstream Protenix checkpoint release version, not part of the public repo id.

Links

• Hub repository: oteam/protenix-base
• Upstream paper: Protenix - Advancing Structure Prediction Through a Comprehensive AlphaFold3 Reproduction
• Upstream repository: Protenix
• Source checkpoint release: https://github.com/bytedance/Protenix/blob/main/docs/supported_models.md
• Code: ofoldx/pipelines/structure_prediction.py
• Project repository: https://github.com/OTeam-AI4S/OFoldX
• Issues: https://github.com/OTeam-AI4S/OFoldX/issues

Usage

The artifact depends on the ofoldx library. Install it with pip:

pip install ofoldx

Pipeline Usage

Load the artifact from oteam/protenix-base with the OFoldX task pipeline. Use AutoModel or AutoProcessor only when you need lower-level control:

from ofoldx.pipelines import Pipeline

pipeline = Pipeline.from_pretrained("oteam/protenix-base")

When a matching processor is available, load it with AutoProcessor.from_pretrained(...) and pass the processed batch to the model.

Interface

• Task: structure_prediction
• Artifact kind: pipeline
• Architecture: alphafold3
• Runtime files: manifest.json, config.json, and model.safetensors when present

Training Details

OFoldX did not train these weights. This repository contains a converted checkpoint and OFoldX runtime metadata for loading it.

Training Data

The upstream Protenix v0.5.0 release reports a 2021-09-30 training-data cutoff. The Protenix training-data pipeline covers released wwPDB/mmCIF bioassemblies, MSAs, CCD cache data, and RCSB sequence clusters. OFoldX does not redistribute the training corpus.

Training Procedure

The upstream protenix_base_default_v0.5.0 model is an AF3-shaped all-atom structure-prediction checkpoint. OFoldX converts protenix_base_default_v0.5.0.pt to the OFoldX artifact format; it does not retrain the model.

Evaluation

OFoldX conversion reports and contract tests validate artifact structure and checkpoint loading. Task-level scientific evaluation should be checked against the corresponding upstream model release or paper.

Limitations

• This artifact is distributed for research use.
• Inputs must match the model-specific processor and expected biomolecular representation.
• OFoldX is pre-alpha, so APIs and artifact metadata may still change before a stable release.

Citation

Please cite the upstream Protenix work for the source checkpoint. If OFoldX supports your work, please also cite or link the OFoldX project repository.

@article{protenix2025advancing,
  author = {ByteDance AML AI4Science Team and Chen, Xinshi and Zhang, Yuxuan and Lu, Chan and Ma, Wenzhi and Guan, Jiaqi and Gong, Chengyue and Yang, Jincai and Zhang, Hanyu and Zhang, Ke and others},
  title = {Protenix - Advancing Structure Prediction Through a Comprehensive AlphaFold3 Reproduction},
  year = {2025},
  doi = {10.1101/2025.01.08.631967},
  journal = {bioRxiv}
}

Contact

Please use OFoldX GitHub issues for questions or comments about this model card.

License

The Hub license metadata, when present, reflects the source checkpoint or upstream project license. The OFoldX project license is not yet finalized. The source checkpoint is associated with the upstream license noted above: Apache-2.0 for upstream Protenix code and model parameters. Review both OFoldX and upstream terms before redistribution or production use.