URSA Reaction Plausibility: leaderboard
The reaction-level validator test of URSA (Insilico Medicine, 2026): models label 1,000 expert-annotated reactions collected from synthesis-planner outputs, half plausible and half implausible, as plausible or not. Accuracy in percent; higher is better.
Metric: Accuracy (%; binary plausible or implausible verdicts on the 1,000 expert-labeled reactions of URSA-reaction-plausibility-bench-2026, 500 plausible and 500 implausible, collected from synthesis-planner outputs; default reasoning effort; higher is better). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 83 | #94 |
| 2 | GPT-5.5 | 76 | #49 |
| 3 | Grok 4.3 | 75 | #333 |
| 4 | Claude Opus 4.8 | 73 | #56 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=ursa-reaction-plausibility · How It Works · Data refreshed daily, snapshot 2026-11-13.