CareQA-Vision - MCQ - Medicine: leaderboard
The 123 image-based multiple-choice questions of CareQA-Vision taken from the Spanish MIR medical residency exams (2020-2024), translated to English. Accuracy in percent, scored by exact match under greedy decoding; higher is better.
Metric: Accuracy (%; 123 image-based multiple-choice questions from MIR medicine exams, exact match). Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 30B A3B Instruct | 76.42 | #870 |
| 2 | Qwen 2 VL 72B Instruct | 75.61 | #1403 |
| 3 | GLM-4.5V | 74.8 | #739 |
| 4 | Qwen 3 VL 8B Instruct | 60.98 | #982 |
| 5 | Qwen 2 VL 7B Instruct | 56.91 | #2569 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=careqa-vision-mcq-medicine · How It Works · Data refreshed daily, snapshot 2026-11-12.