reasoning benchmark
GeneBench-Pro
GeneBench-Pro is a research-level benchmark of 129 multi-stage computational-biology problems spanning genomics, quantitative biology, and translational biomedicine. Each problem gives the agent a messy dataset, brief context, and a target estimand, and requires navigating dependent inferential decision points to reach a verifiable answer.
Updated Aug 7, 2026
Published models3
Registry coverage3
MetricScore
EvidenceB
GeneBench-Pro leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | OP | 28.7% | 100.0% | 3 | C | |
| 2 | OP | 23.3% | 50.0% | 3 | C | |
| 3 | OP | 10.8% | 0.0% | 3 | C |
Score distribution
Top published rows on the benchmark's original scale.
GeneBench-Pro highlights
The top published results on this benchmark's own scale.
What is GeneBench-Pro?
Definition and scoring fields from the benchmark registry.
GeneBench-Pro is a research-level benchmark of 129 multi-stage computational-biology problems spanning genomics, quantitative biology, and translational biomedicine. Each problem gives the agent a messy dataset, brief context, and a target estimand, and requires navigating dependent inferential decision points to reach a verifiable answer.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- GeneBench-Pro
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- genebench-pro|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about GeneBench-Pro.
Which model scores highest on GeneBench-Pro?
GPT-5.6 Sol is currently ranked first with 28.7%.
What does GeneBench-Pro measure?
GeneBench-Pro is a research-level benchmark of 129 multi-stage computational-biology problems spanning genomics, quantitative biology, and translational biomedicine. Each problem gives the agent a messy dataset, brief context, and a target estimand, and requires navigating dependent inferential decision points to reach a verifiable answer.
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.