Protenix (AlphaFold3)

Another AlphaFold3 implementation developed by the ByteDance team.

Protenix (AlphaFold3) media

Protenix is an open-source reproduction of AlphaFold3, enabling researchers to predict 3D structures of biomolecular complexes. It incorporates innovations in model architecture, data processing, and confidence prediction, democratizing access to advanced tools for modeling biomolecular interactions.

Estimated resource cost: 0.0024025

Categories: Structure Prediction & Folding

Tags: Co-Folding Conformation Generation Nucleotides Protein Folding Proteins Protein–Ligand Docking Protein–Protein Docking Small Molecules

Key Capabilities

  • Reproduces AlphaFold3's architecture for predicting 3D structures of biomolecular complexes.
  • Supports diverse biomolecular systems, including proteins, nucleic acids, and small molecules.
  • Employs innovative algorithms for MSA pairing, pocket-conditioning, and unified cropping.
  • Optimized confidence model for reliable biomolecular interaction predictions.
  • Significant speed and computational efficiency improvements over comparable models.
  • Extensive benchmark validation on CASP15 and curated test sets.

Runtime Statistics

MetricValue
runtime_mean1362
runtime_median375
runtime_std3449
runtime_90th_percentile3132
runtime_max50778

Similar Tools

  • Boltz-1 (AlphaFold3)
  • Boltz-2 (AlphaFold3)
  • Chai-1 (AlphaFold3)
  • IntelliFold (AlphaFold3)
Job submission preview

Ready to submit your job?

Review your configuration, then confirm the estimated credit cost before you run the job. Note that credit estimates are not guaranteed and runtime can vary depending on inputs and settings.

Estimated Credits: 0.0024025

Invite-only, limited-time access. Please contact ztang@getantibody.com.

Invite-only, limited-time access. Please contact ztang@getantibody.com.

Back to Service Catalog Back to Catalog