CareTransition-Audit - Accuracy: leaderboard
Metric: Label accuracy (%) against the clinician labels on the 46-question DISCHARGED audit checklist applied to 50 MIMIC-IV discharge summaries (Yes, No, Unclear or N/A per question), zero-shot with six prompts per summary and an indirect chain-of-thought instruction; summary-question pairs with an N/A label or an unparseable output are left out; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 81.4 |
| 2 | Claude Sonnet 4.5 | 80.4 |
| 3 | GPT-5.4 | 77.2 |
| 4 | Nova 2.0 Lite (Non-reasoning) | 74.7 |
| 5 | DeepSeek V3.2 | 74.3 |
| 6 | GPT-4o | 72.8 |
| 7 | Grok 4.1 Fast | 72.1 |
| 8 | Llama 4 Maverick | 70.6 |
| 9 | Phi-4 | 64 |
| 10 | Qwen 2.5 7B Instruct | 62.3 |
Interactive version: theaggregate.ai/benchmark?slug=caretransition-audit-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-07.