CareTransition-Audit - Completeness Ranking: leaderboard

Discharge-summary auditing on CareTransition-Audit (2026): the model answers the 46 checklist questions of the DISCHARGED framework (Yes, No, Unclear or N/A) for each of 50 MIMIC-IV discharge summaries, and its labels are compared with clinician labels. Spearman correlation from -1 to 1 between the per-summary documentation completeness (share of Yes labels) assigned by the model and by the clinician; higher is better.

Metric: Spearman correlation (-1 to 1) between the per-summary documentation completeness (share of Yes labels) the model assigns and the one the clinician assigns, over the 50 MIMIC-IV discharge summaries of the 46-question DISCHARGED audit checklist; higher is better. Source: arxiv.org. Saturation forecast: Around March 2027. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Gemini 3 Flash (Preview)0.537#152
2Llama-3.3-nemotron-super-49B-v1.50.474#920
3Claude Sonnet 4.50.432#261
4DeepSeek V3.20.413#419
5GPT-5.40.365#135
6Nova 2.0 LiteNon-reasoning0.359#1177
7GPT-4o0.341#751
8Llama 4 Maverick0.335#1110
9Grok 4.1 Fast0.263#427
10Qwen 2.5 7B Instruct0.254#2723
11Phi-40.1#1931

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=caretransition-audit-completeness-ranking · How It Works · Data refreshed daily, snapshot 2026-11-21.