AgentEHR (ReAct) - Labevents: leaderboard
Metric: F1 (%; predicted labevents label set against the next recorded events, 589 MIMIC-IV-Common cases; standard ReAct agent). Source: arxiv.org. Saturation forecast: Around March 2028. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 Mini | 21.51 |
| 2 | Grok 4.1 Fast | 18.56 |
| 3 | Qwen 3 30B A3B | 12.95 |
| 4 | Qwen 3 Next 80B A3B | 7.93 |
Interactive version: theaggregate.ai/benchmark?slug=agentehr-react-labevents · How It Works · Data refreshed daily, snapshot 2026-09-26.