Pocket-Dentist - BRAR Bone Resorption Grading: leaderboard
Metric: Macro-F1 (%; three bone-resorption grades from a panoramic radiograph and patient metadata; 149 test radiographs; zero-shot). Source: arxiv.org. Saturation forecast: Around December 2026. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Flash | 41.1 |
| 2 | Lingshu-32B | 39.2 |
| 3 | MedGemma-4B-IT | 33 |
| 4 | Gemma 4 E2B (IT) | 24 |
| 5 | GPT-4o Mini | 23.9 |
| 6 | gemma-4-E4B-it | 23.9 |
| 7 | Qwen 2.5 VL 7B Instruct | 16 |
Interactive version: theaggregate.ai/benchmark?slug=pocket-dentist-brar-bone-resorption-grading · How It Works · Data refreshed daily, snapshot 2026-09-26.