MMRareBench - Cross-Image Evidence Alignment: leaderboard
MMRareBench (2026) cross-image track: the model describes findings in two or more medical images of different modalities from a rare-disease case, aligns the evidence across them and names their relation. Rubric score (0-100) from a strict Qwen3-VL-235B judge; higher is better.
Metric: Cross-Image Evidence Alignment track (T3) score: per-image findings, their alignment across two or more imaging modalities and the relation type, 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 (five binary dimensions averaged); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 23 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 70.8 | #178 |
| 2 | Gemini 2.5 Pro | 66.9 | #276 |
| 3 | Gemini 3 Flash (Preview) | 62.9 | #152 |
| 4 | Qwen 3 VL 235B A22B Instruct | 56.6 | #532 |
| 5 | Gemini 2.5 Flash | 54.9 | #463 |
| 6 | Claude Sonnet 4.5 | 54.7 | #261 |
| 7 | Claude Haiku 4.5 | 41.9 | #583 |
| 8 | Qwen 3 VL 30B A3B Instruct | 36.9 | #895 |
| 9 | GLM-4.6V | 28.7 | #710 |
| 10 | MedGemma-27B-IT | 27.3 | #1350 |
| 11 | Qwen 3 VL 8B Instruct | 25.7 | #1029 |
| 12 | Qwen 2.5 VL 32B Instruct | 21.2 | #1062 |
| 13 | Qwen 2.5 VL 72B Instruct | 18.9 | #841 |
| 14 | GPT-4o | 15.6 | #751 |
| 15 | Lingshu-32B | 8.7 | #1113 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=mmrarebench-cross-image-evidence-alignment · How It Works · Data refreshed daily, snapshot 2026-11-21.