Protein Structure · Chai Discovery
Open multimodal complex prediction across proteins, nucleic acids and ligands.
5–30min
Runs on On-demand GPU
Runs on our servers with on-demand compute. A first run needs time to load the model; active capacity can be reused and scales down when idle.Run this model
About this model
Chai-1 is the credible open alternative to AlphaFold3 for multi-entity complexes. It handles protein–protein, protein–nucleic-acid and protein–ligand assemblies in one model, with open weights for research use.
Standardized I/O contract
Every model in the Hub speaks the same contract, which is what lets the Router and the agent call any of them without special-casing.
Inputs
Entity sequences
Search for homologues (MSA)
Recycles
Structures to sample
Outputs
All-atom coordinates in PDB format, viewable and downloadable.
pLDDT score per residue (0–100). Above 70 is generally reliable.
