LoMeVQA - Progress Classification: leaderboard
Metric: Accuracy (%; three-option classification of disease progression between two radiographs of the same patient; LoMeVQA-test, 2,500 expert-reviewed longitudinal chest radiograph samples over five tasks, zero-shot; higher is better). Source: arxiv.org. Saturation forecast: Around 2032. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Lingshu-32B | 54.8 |
| 2 | MedGemma-4B | 52.6 |
| 3 | Lingshu-7B | 52 |
| 4 | GPT-5 | 47.4 |
| 5 | Qwen 3 VL 32B | 46.4 |
| 6 | MedGemma-27B-IT | 45.8 |
| 7 | Gemini 2.5 Flash | 44.4 |
| 8 | GPT-4o | 38.4 |
| 9 | InternVL3.5-8B | 32.4 |
Interactive version: theaggregate.ai/benchmark?slug=lomevqa-progress-classification · How It Works · Data refreshed daily, snapshot 2026-09-29.