Why this evidence?
It is linked to a specific model version, scientific task, dataset, split, metric and source. That makes the claim inspectable rather than a detached marketing score.
Chai-1 / Chai-2: Biomolecular complex prediction
Developer-reported AF3-class complex-prediction performance.
Model versionChai-2
TaskBiomolecular complex prediction
DatasetComplex and antibody-design evaluations
SplitPoseBusters and complex-evaluation sets
MetricStructure / design performance
Replicationpartial
Reported byModel developers
Review statuscurated
Evidence confidence · limited
Confidence is multidimensional, not a universal model score.
Evidence completenessstrongModel version, task, dataset, split, metric, source and provenance fields.
Independent validationmoderatepartial
Source qualitylimitedPreprint / developer-reported
Reproducibility evidencemoderateReflects documented replication status, not a universal reproducibility score.
Version specificitystrongChai-2
Context applicabilitystrongDepends on task, split and explicit caveats; users must still validate their own context.
Contradiction reviewclearNo direct contradiction signal is currently queued.
BioAtlas reports evidence dimensions separately so a strong source cannot hide weak applicability, incomplete replication or unresolved contradiction.
Why?
Why should this evidence influence a decision?
Why not a universal score?
Performance can change with dataset, split, preprocessing, metric and context of use. BioAtlas therefore keeps confidence dimensions separate.
What could change the conclusion?
Independent replication, a better matched prospective dataset, a version change, a contradictory result or a more relevant validation protocol can reopen this evidence record.
Evidence boundary
What this claim does not prove.
- Preprint and developer-reported comparisons require independent reproduction.
- Benchmark protocol and entity coverage determine comparability.
BioAtlas groups benchmark claims only when task, dataset, split, metric and protocol context align. This record is not a universal model score.
Model context
Chai-1 is a multi-modal biomolecular structure model on par with AlphaFold 3 for predicting proteins with ligands, nucleic acids and multimers. Chai Discovery (backed by OpenAI and Thrive) released it free for research and drug discovery, and followed with Chai-2 for zero-shot antibody design.
Full evidence passport →Known model limitations
- Performance depends on the evaluation dataset and operating conditions.
- Task-specific benchmark results should not be compared across unlike domains.
- Outputs require task-specific scientific and experimental validation.