MMRareBench - Examination Suggestion: leaderboard
Metric: Examination Suggestion track (T4) score: prioritized next-step investigations for a case narrative truncated before the gold tests, graded 0-100 by a Qwen3-VL-235B judge with a strict track-specific rubric that awards credit only when a criterion is fully met (eight 0-2 dimensions normalized by 16); higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 | 85.9 |
| 2 | Gemini 2.5 Pro | 85.5 |
| 3 | Qwen 3 VL 235B A22B Instruct | 83.2 |
| 4 | Gemini 3 Flash (Preview) | 81.8 |
| 5 | Claude Haiku 4.5 | 80.6 |
| 6 | GPT-5 | 77.9 |
| 7 | Gemini 2.5 Flash | 66.9 |
| 8 | Qwen 3 VL 30B A3B Instruct | 63.1 |
| 9 | MedGemma-27B-IT | 56 |
| 10 | Qwen 3 VL 8B Instruct | 43.9 |
| 11 | GPT-4o | 42.8 |
| 12 | Qwen 2.5 VL 72B Instruct | 42.3 |
| 13 | Qwen 2.5 VL 32B Instruct | 42 |
| 14 | Lingshu-32B | 36.9 |
| 15 | MedGemma-4B-IT | 33.1 |
Interactive version: theaggregate.ai/benchmark?slug=mmrarebench-examination-suggestion · How It Works · Data refreshed daily, snapshot 2026-10-07.