Synthetic Hospital - Summarization: leaderboard
Synthetic Hospital (Park, Chen and Dettmers, 2026) is an open synthetic longitudinal EHR benchmark built from medical-education material, with labels grounded in ICD-10-CM, SNOMED CT and LOINC. In this task the model writes a structured whole-patient clinical summary, measured by finding-level F1 in percent against key findings, scored against graph-derived references on the 200-patient public split with one prompting strategy locked per task; higher is better.
Metric: Finding-level F1 (%; whole-patient clinical summary scored against graph-defined key findings, ontology-grounded structured prompting; 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 August 2028. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 55 | #100 |
| 2 | Kimi K2.5 (Thinking) | 53.2 | #252 (Kimi K2.5) |
| 3 | GPT-5.3 | 48.9 | #194 |
| 4 | GLM-5 | 48.4 | #256 |
| 5 | Qwen 3.5 397B A17B | 42.5 | #242 |
| 6 | Llama 4 Scout | 40.3 | #1787 |
| 7 | DeepSeek V3.2 | 39.9 | #403 |
| 8 | Gemma 3 27B | 38.5 | #1638 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=synthetic-hospital-summarization · How It Works · Data refreshed daily, snapshot 2026-11-12.