CLIR-Bench - Intervention Decision: leaderboard
Metric: Accuracy (%) on immediate intervention decision, 4-option multiple-choice questions over serialized irregular ICU time series (Full-TS input), chance 25; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2.6 | 91.67 |
| 2 | GPT-5.4 Mini | 76.67 |
| 3 | Gemini 2.5 Flash | 60 |
| 4 | DeepSeek V4 Flash | 41.67 |
| 5 | Gemma 3 27B (IT) | 39.67 |
| 6 | Gemma 4 12B (IT) | 39.67 |
| 7 | Qwen 3.5 4B | 22 |
| 8 | Qwen 3.5 9B | 20 |
| 9 | Qwen 3.6 27B | 14.33 |
Interactive version: theaggregate.ai/benchmark?slug=clir-bench-intervention-decision · How It Works · Data refreshed daily, snapshot 2026-09-29.