Model type
Multimodal biomolecular structure predictor
Configuration
Chai-1 (single sequence)
Chai-1 predicts the structures of biomolecular complexes containing proteins, nucleic acids and small molecules.
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README LicenceHow it worksChai-1 workflow
Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README LicenceOverview
Inputs
FASTA sequences, ligand SMILES and optional alignments, templates, contacts or covalent-bond restraints.
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README LicenceOutputs
Sampled complex structures; the default command produces five predictions.
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README LicenceAccess
Official project documentation and implementation: https://github.com/chaidiscovery/chai-lab
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licencelimited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
1 evaluation · 1 metric rows. Different protocols are not a single leaderboard.
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: Chai-1 (single sequence) | Protocol: ESMFold2 Runs N’ Poses reported comparison single-sequence: Runs N’ Poses ligand pass rate (single sequence) Dataset subset: Runs N’ Poses complete-case intersection: 2,573 scored ligands (ESMFold2 Runs N’ Poses reported comparison split) | 51% ligand_pass_rate percent · higher Uncertainty: Not reported Coverage: unit: ligands; scored: 2573; eligible: unreported; note: Complete-case intersection; source 2600 systems is not a ligand denominator. | Author-reported evaluation · source checkedMethods, coverage and sourceRuns N’ Poses receptor–ligand co-folding; source benchmark 2,600 systems. Figure 2C reports n=2,573 scored ligands on the intersection where all models produced valid predictions, after excluding undefined SuCOS scores. Multiple ligands in one system are scored independently. Five seeds × five diffusion samples per target; select top candidate by ipTM. Success requires lDDT-PLI >0.8 and BiSyRMSD <2 angstrom. Baselines use 10 recycles and 200 diffusion steps; ESMFold2 uses 10 or 20 loops as labelled and truncated 68-step diffusion. Single-sequence and MSA conditions remain separate. Aggregation: Not reported ESMFold2 primary paper v1, Figure 2C Runs N’ Poses · PDF page 5, Figure 2C, Runs N’ Poses subpanel (right), single-sequence block, bar 1 from left (Chai-1 (single sequence)), exact printed bar label |
Source checking is not independent reproduction. Release 2026-09-23-2b89723c6dd9.
Use this model
How it works, versions and access
Related profile: Chai-1. This page retains the exact record and its evaluation context.
This configuration
Author-evaluated folding configuration with conditioning and loop count retained from Figure 2C.
- record
- Chai-1 (single sequence)
- configuration
- Not reported
- entity type
- Configuration
How it works
How it works
Chai-1 predicts the structures of biomolecular complexes containing proteins, nucleic acids and small molecules. An AlphaFold3-like structure architecture dominated by pair-biased self-attention, with additional protein-language-model embeddings and optional inter-chain constraint features. The documented inputs are FASTA sequences, ligand SMILES and optional alignments, templates, contacts or covalent-bond restraints. The output consists of sampled complex structures; the default command produces five predictions.
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README LicenceVersions and reproducibility
Chai-1; README installation example pins chai_lab 0.6.1. The applicable input limits require configuration-specific checking.
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README LicenceStrengths, limitations and unresolved questions
Strengths and limitations
Strengths and considerations
- Supports user-provided restraints and covalent bonds, alongside MSA and template inputs.
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence
Limitations and conditions
- The report documents failures of relative chain placement and sensitivity to modified residues. Inputs, MSA/template evidence and sampled structures must remain explicit when comparing runs.
Sources (2)
chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence
Profile review details
Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.
Stable record: catalog-model-chai-1Specifications
Inputs, training, access and other details
Explanatory profile: limited source coverage · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.
| Property | Description and evidence |
|---|---|
| Model type | Multimodal biomolecular structure predictorSources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Architecture | An AlphaFold3-like structure architecture dominated by pair-biased self-attention, with additional protein-language-model embeddings and optional inter-chain constraint features.Sources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Inputs | FASTA sequences, ligand SMILES and optional alignments, templates, contacts or covalent-bond restraints.Sources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Outputs | Sampled complex structures; the default command produces five predictions.Sources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Parameters | The September 2024 report specifies a 3B protein-language-model component but does not state a complete predictor total in the reviewed architecture sections. · Not reported in inspected sourcesSources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Known versions | Chai-1; README installation example pins chai_lab 0.6.1.Sources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Training data | PDB structures and AlphaFoldDB distillation, with optional MSAs/templates. The technical report describes no other AlphaFold3 distillation datasets; server MSA search differs from the paper evaluation pipeline.Sources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Training cutoff | PDB structure and PDB70-template release cutoff: 2021-01-12, according to the September 2024 technical report.Sources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Context limits | The reviewed report and inference README do not specify one validated maximum for all protein, nucleic-acid and ligand inputs; resource requirements and molecular composition remain relevant. · Not reported in inspected sourcesSources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Weights licence | Apache-2.0; README Licence explicitly covers code and weights.Sources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Access | Official project documentation and implementation: https://github.com/chaidiscovery/chai-labSources (2)chaidiscovery/chai-lab: README.md; chai1-web: Browser-extracted primary-paper passages · Chai-1 Technical Report v1 (9 September 2024), Sections 2.1, 2.7–2.8 and 4.1–4.3; current repository README Licence |
| Code licence | Apache-2.0Sourceschaidiscovery/chai-lab: LICENSE · LICENSE: licence text |
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Relationship: family catalog-model-chai-1 Individual claims | ESMFold2 primary paper v1, Figure 2C Runs N’ Poses Figure 2C Runs N’ Poses, bar label Chai-1 (single sequence); Appendix A.2.10 Version: 10.64898/2026.06.03.729735v1; posted 2026-06-04 | source checked automated source review · 2026-09-23 Audit detailsSource-backed evaluated identity only; no independent reproduction. Field: Claim: esmfold2-2026-runs-n-poses-method-chai-1-single-sequence-catalog-model-chai-1-identity-claim Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
View linked audit checks and correction history
Release 2026-09-23-2b89723c6dd9 · Record review: source checked
1 source records and release history
- ESMFold2 primary paper v1, Figure 2C Runs N’ Poses · Original source · 10.64898/2026.06.03.729735v1; posted 2026-06-04
Technical metadata and extraction receipts
Stable ID: esmfold2-2026-runs-n-poses-method-chai-1-single-sequence
- areas
- proteins-complexes
- source locator
- Figure 2C Runs N’ Poses, bar label Chai-1 (single sequence); Appendix A.2.10
- missing metadata
- checkpoint revision: unreported; parameters: unextracted