Model type
Biomolecular structure predictor; this record is the paper-specific evaluated configuration.
Configuration
Chai-1
Chai-1 is a general structure-prediction comparator on the Ibex immunoglobulin benchmark.
Sources
Conformation-aware structure prediction of antigen-recognizing immune proteins · Discussion (paragraph 6); Results/Benchmarking on the ImmuneBuilder test set (paragraph 1)How it worksEvaluated procedure (conceptual)
Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings.
Sources
Conformation-aware structure prediction of antigen-recognizing immune proteins · Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); Results/Benchmarking on a private dataset (paragraph 3)At a glance
Biological inputs
Antibody, nanobody or T-cell-receptor sequences
Sources
Conformation-aware structure prediction of antigen-recognizing immune proteins · Introduction (paragraph 5); Results/Benchmarking on the ImmuneBuilder test set (paragraph 2)Outputs
Predicted structures
Sources
Conformation-aware structure prediction of antigen-recognizing immune proteins · Methods/Training (paragraph 1); The Ibex model (paragraph 2)limited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
Release 2026-09-17-d277315f7d76 · 1 evaluation · 1 metric row. Different protocols are not a single leaderboard.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| Chai-1: Antibody loop structure prediction Configuration: Chai-1Task: Antibody loop structure predictionDataset: ImmuneBuilder antibody test set Backbone RMSD after framework alignment; one seed and one diffusion trajectory. Independent external evaluation · Evaluation metadata: needs review | ||
| 2.65 Mean CDR H3 RMSD Unit: Å · Direction: unknown | Uncertainty: not reported in legacy extract Scored: Not reported · Eligible: Not reported | source checkedConformation-aware structure prediction of antigen-recognizing immune proteins · Table 1, Antibodies / Chai-1 row, CDR H3 column Source checking is not independent reproduction. |
How it works
How the evaluated method works
The study uses one seed and one diffusion trajectory per prediction, then aligns frameworks and measures region-specific backbone RMSD.
Sources
Conformation-aware structure prediction of antigen-recognizing immune proteins · Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); Results/Benchmarking on a private dataset (paragraph 3)Underlying method and version boundaries
Chai-1 predicts biomolecular structures through the released inference implementation. Its input options, MSA/template use and sampling budget are part of each evaluated configuration.
Sources
github.com/chaidiscovery/chai-lab README.md · README.md; introduction, model description, pretrained-model and usage sections at pinned revisionWhat was evaluated
The linked evaluation record identifies Chai-1: Antibody loop structure prediction. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
Conformation-aware structure prediction of antigen-recognizing immune proteins · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-b3-041Strengths and limitations
Strengths and considerations
- Provides a general biomolecular predictor alongside specialised immunoglobulin methods.
Sources
Conformation-aware structure prediction of antigen-recognizing immune proteins · Discussion (paragraph 6); Discussion (paragraph 3)
Limitations and conditions
- The paper reports different behaviour across loop types and modalities; the single-sample configuration is not a maximum-sampling capability estimate.
Sources
Conformation-aware structure prediction of antigen-recognizing immune proteins · Methods/Training (paragraph 2); Results/Benchmarking on the ImmuneBuilder test set (paragraph 2)
Profile review details
Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction.
Stable record: reported-model-70c732770a200fSpecifications
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 | Biomolecular structure predictor; this record is the paper-specific evaluated configuration.Sourcesgithub.com/chaidiscovery/chai-lab README.md · README.md model description |
| Architecture / procedure | The study uses one seed and one diffusion trajectory per prediction, then aligns frameworks and measures region-specific backbone RMSD.SourcesConformation-aware structure prediction of antigen-recognizing immune proteins · Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); Results/Benchmarking on a private dataset (paragraph 3) |
| Biological inputs | Antibody, nanobody or T-cell-receptor sequencesSourcesConformation-aware structure prediction of antigen-recognizing immune proteins · Introduction (paragraph 5); Results/Benchmarking on the ImmuneBuilder test set (paragraph 2) |
| Outputs | Predicted structuresSourcesConformation-aware structure prediction of antigen-recognizing immune proteins · Methods/Training (paragraph 1); The Ibex model (paragraph 2) |
| Parameters | An aggregate parameter total for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sourcesSources (2)Conformation-aware structure prediction of antigen-recognizing immune proteins; github.com/chaidiscovery/chai-lab README.md · Methods/Model; Methods/Data; Methods/Training; Methods/Private dataset generation and characterization; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.md at pinned repository revision |
| Known versions / configuration | One seed and one diffusion trajectory in the ImmuneBuilder comparison · Not reported in inspected sourcesSourcesConformation-aware structure prediction of antigen-recognizing immune proteins · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. |
| Training data / fitting | The study evaluates the released structure predictor on the ImmuneBuilder test set using one seed and one diffusion trajectory. It does not report task-specific retraining of Chai-1; a complete upstream training inventory is not restated.SourcesConformation-aware structure prediction of antigen-recognizing immune proteins · Results / Benchmarking on the ImmuneBuilder test set |
| Context limits | A maximum input/context length for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sourcesSources (2)Conformation-aware structure prediction of antigen-recognizing immune proteins; github.com/chaidiscovery/chai-lab README.md · Methods/Model; Methods/Data; Methods/Training; Methods/Private dataset generation and characterization; inspected for explicit maximum input length (dataset lengths and family-wide limits are not substituted); README.md at pinned repository revision |
| Access | Official upstream implementation and usage documentation: https://github.com/chaidiscovery/chai-lab/blob/66c38d1fe5c6756a89ff8596b1dea87d305ec06f/README.md. This pinned documentation revision is not automatically the evaluated weight revision.Sourcesgithub.com/chaidiscovery/chai-lab README.md · README.md; installation, model download and usage instructions |
| Code licence | Apache 2.0 (upstream repository code at the cited revision; this does not establish every dependency or historical checkpoint licence).Sourcesgithub.com/chaidiscovery/chai-lab LICENSE · LICENSE; complete licence text |
| Weights licence | Apache 2.0 for both Chai-1 code and model weights, explicitly stated in the official README; paper-specific derived artifacts are separate.Sourcesgithub.com/chaidiscovery/chai-lab README.md · README.md; Licence section |
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.
22 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings. Individual claims | Conformation-aware structure prediction of antigen-recognizing immune proteins Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); Results/Benchmarking on a private dataset (paragraph 3) Version: PMC archival version PMC12710905.1 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram steps ["Antibody, nanobody or T-cell-receptor sequences","Chai-1","Predicted structures"] Individual claims | Conformation-aware structure prediction of antigen-recognizing immune proteins Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); Results/Benchmarking on a private dataset (paragraph 3) Version: PMC archival version PMC12710905.1 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Evaluated procedure (conceptual) Individual claims | Conformation-aware structure prediction of antigen-recognizing immune proteins Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); Results/Benchmarking on a private dataset (paragraph 3) Version: PMC archival version PMC12710905.1 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Model type Biomolecular structure predictor; this record is the paper-specific evaluated configuration. Individual claims | github.com/chaidiscovery/chai-lab README.md README.md model description Version: 66c38d1fe5c6756a89ff8596b1dea87d305ec06f | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Architecture / procedure The study uses one seed and one diffusion trajectory per prediction, then aligns frameworks and measures region-specific backbone RMSD. Individual claims | Conformation-aware structure prediction of antigen-recognizing immune proteins Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); Results/Benchmarking on a private dataset (paragraph 3) Version: PMC archival version PMC12710905.1 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Weights licence Apache 2.0 for both Chai-1 code and model weights, explicitly stated in the official README; paper-specific derived artifacts are separate. Individual claims | github.com/chaidiscovery/chai-lab README.md README.md; Licence section Version: 66c38d1fe5c6756a89ff8596b1dea87d305ec06f | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Biological inputs Antibody, nanobody or T-cell-receptor sequences Individual claims | Conformation-aware structure prediction of antigen-recognizing immune proteins Introduction (paragraph 5); Results/Benchmarking on the ImmuneBuilder test set (paragraph 2) Version: PMC archival version PMC12710905.1 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Outputs Predicted structures Individual claims | Conformation-aware structure prediction of antigen-recognizing immune proteins Methods/Training (paragraph 1); The Ibex model (paragraph 2) Version: PMC archival version PMC12710905.1 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Parameters An aggregate parameter total for this exact evaluated configuration is not established by the inspected sources. Individual claims | github.com/chaidiscovery/chai-lab README.md Methods/Model; Methods/Data; Methods/Training; Methods/Private dataset generation and characterization; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.md at pinned repository revision Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 66c38d1fe5c6756a89ff8596b1dea87d305ec06f | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Parameters An aggregate parameter total for this exact evaluated configuration is not established by the inspected sources. Individual claims | Conformation-aware structure prediction of antigen-recognizing immune proteins Methods/Model; Methods/Data; Methods/Training; Methods/Private dataset generation and characterization; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.md at pinned repository revision Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: PMC archival version PMC12710905.1 | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
Release 2026-09-17-d277315f7d76 · Record review: needs review
3 source records and release history
- Conformation-aware structure prediction of antigen-recognizing immune proteins · Original source · PMC archival version PMC12710905.1
- github.com/chaidiscovery/chai-lab README.md · Original source · 66c38d1fe5c6756a89ff8596b1dea87d305ec06f
- github.com/chaidiscovery/chai-lab LICENSE · Original source · 66c38d1fe5c6756a89ff8596b1dea87d305ec06f
Technical metadata and extraction receipts
Stable ID: reported-model-70c732770a200f
- areas
- molecular-interactions
- entity level
- method
- version
- Not reported
- reported name
- Chai-1
- historical missing metadata
- version: not_reported_in_legacy_extract; checkpoint revision: not_reported_in_legacy_extract; training data: not_reported_in_legacy_extract; licence: not_reported_in_legacy_extract
- metadata review scope
- historical_missing_metadata preserves the original discovery state. Current descriptive evidence and missingness are recorded in profile.facts; numerical-result review is separate.
- legacy kinds
- model
- entity classification
- review date: 2026-09-17; rationale: This source-scoped entry preserves the method/configuration actually named in an evaluation. It is neither a global family identity nor proof of an immutable checkpoint; the linked evaluation retains adaptation, fitting and scoring details.; source ids: ibex-2025; evidence-reported-base-chai-readme-md; source locator: Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); Results/Benchmarking on a private dataset (paragraph 3) | README.md model description | Discussion (paragraph 6); Results/Benchmarking on the ImmuneBuilder test set (paragraph 1); ambiguities: Configuration means the source-labelled evaluated identity. It does not establish missing checkpoint hashes, default settings or equivalence to same-named records in other papers.