MedLayXPlain: leaderboard
Medical Vision-Language Models (Med-VLMs) achieve strong expert-level performance, yet their ability to generate patient-accessible descriptions remains underexplored. With the 21st Century Cures Act now mandating immediate patient access to diagnostic imaging results, evaluating whether Med-VLMs can bridge this Expert-Lay Gap is both urgent and clinically consequential for patient education and s
Metric: S (self-reported). Source: benchmarklist.com. Saturation forecast: Not forecast. 32 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash (Thinking) | 70.6 | #245 (Gemini 2.5 Flash) |
| 2 | Qwen 3 VL 32B (Thinking) | 65.4 | #294 (Qwen 3 VL 32B) |
| 3 | Qwen 3 VL 30B A3B (Thinking) | 64.8 | #351 (Qwen 3 VL 30B A3B) |
| 4 | InternVL3-8B | 64.7 | #620 |
| 5 | Qwen 3 VL 8B (Thinking) | 64.6 | |
| 6 | Claude Opus 4.7 | 64.1 | #49 |
| 7 | Qwen 3.5 2B (Thinking) | 63.7 | #1148 (Qwen 3.5 2B) |
| 8 | Qwen 3 VL 4B (Thinking) | 63.3 | #480 (Qwen 3 VL 4B) |
| 9 | Qwen 3.6 35B A3B | 61.5 | #226 |
| 10 | GPT-5.4 Mini | 61.2 | #212 |
| 11 | GPT-5.5 | 60.6 | #31 |
| 12 | Qwen 3.5 0.8B (Thinking) | 58.2 | #1424 (Qwen 3.5 0.8B) |
| 13 | Qwen 3.5 9B (Thinking) | 54.5 | #379 (Qwen 3.5 9B) |
| 14 | Qwen 2.5 VL 7B | 52.1 | #643 |
| 15 | Llama 4 Scout | 51.1 | #656 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medlayxplain · How It Works · Data refreshed daily, snapshot 2026-10-20.