safety benchmark
FigQA
FigQA is a multiple-choice benchmark on interpreting scientific figures from biology papers. It evaluates dual-use biological knowledge and multimodal reasoning relevant to bioweapons development.
Updated Aug 7, 2026
Published models3
Registry coverage3
MetricScore
EvidenceB
FigQA leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AN | 89.0% | 100.0% | 3 | C | |
| 2 | AN | 78.3% | 50.0% | 3 | C | |
| 3 | XA | 34.0% | 0.0% | 3 | C |
Score distribution
Top published rows on the benchmark's original scale.
FigQA highlights
The top published results on this benchmark's own scale.
What is FigQA?
Definition and scoring fields from the benchmark registry.
FigQA is a multiple-choice benchmark on interpreting scientific figures from biology papers. It evaluates dual-use biological knowledge and multimodal reasoning relevant to bioweapons development.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- FigQA
- Modality
- multimodal
- Primary category
- safety
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- figqa|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about FigQA.
Which model scores highest on FigQA?
Claude Mythos Preview is currently ranked first with 89.0%.
What does FigQA measure?
FigQA is a multiple-choice benchmark on interpreting scientific figures from biology papers. It evaluates dual-use biological knowledge and multimodal reasoning relevant to bioweapons development.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
3 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.