ClinBench Quarterly (Insilico)
Quarterly-refreshed clinical-trial outcome benchmark on ScienceAIBench / InsilicoBench / DDB (25 tasks).
Composite
81.5
Experimental validation
Clinical
Stages
Phase IIPhase IIIClinical Development
Modalities
cross-modality
Task types
trial-outcome-prediction
Size
tasks: 25
refresh_cadence_months: 3
refresh_cadence_months: 3
License
CC-BY
First release
2025-01
Last updated
2026-04
Official site
Leaderboard
Dataset
→ dataset
Code / GitHub
→ repository
HuggingFace
→ HF
Paper
ClinBench methodology note · I, n, s, i, l, i, c, o, , A, I, , T, e, a, m · 2025 · paper · doi:N/A — portal · 10 citations
Flags
none
Experts
Groups
Hosted by
Related benchmarks
Rubric (7-criterion)
rigor
4
coverage
3
maintenance
5
adoption
3
quality
4
accessibility
5
industry_relevance
5
Notes
Benchmark refresh cadence beats all academic trial outcome benchmarks. Leaderboards test frontier LLMs against quarterly-updated splits.