Synthetic Hospital - Evidence Retrieval: leaderboard
Metric: Precision at 5 (%; ranking chart sections for a diagnosis against graded graph relevance, zero-shot; public split of 200 synthetic longitudinal patients (1,268 in all, built from USMLE-style questions with ontology-grounded labels); single-turn, one prompting strategy locked per task for every model; graph-derived deterministic scoring). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.3 | 83.3 |
| 2 | Claude Opus 4.6 | 81.6 |
| 3 | Kimi K2.5 (Thinking) | 81.1 |
| 4 | Qwen 3.5 397B A17B | 80.9 |
| 5 | Gemma 3 27B | 80.4 |
| 6 | GLM-5 | 79.8 |
| 7 | DeepSeek V3.2 | 75.1 |
| 8 | Llama 4 Scout | 74.1 |
Interactive version: theaggregate.ai/benchmark?slug=synthetic-hospital-evidence-retrieval · How It Works · Data refreshed daily, snapshot 2026-09-26.