ClinTraceBench - Treatment Fact Recall: leaderboard
Metric: Accuracy (%; task T7, treatment fact recall: four-option clinical-management questions grounded in documented treatment events, 400 questions; full-context strategy: the whole multi-visit patient-clinician dialogue derived from MIMIC-IV records is given verbatim, models called through OpenRouter). Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Haiku 4.5 | 96.5 |
| 2 | Claude Sonnet 4.6 | 95.3 |
| 3 | DeepSeek V3.1 | 94.2 |
| 4 | GPT-4o Mini | 87.2 |
Interactive version: theaggregate.ai/benchmark?slug=clintracebench-treatment-fact-recall · How It Works · Data refreshed daily, snapshot 2026-09-26.