Dental-TriageBench - Coarse-Grained Exact Match: leaderboard
Dental-TriageBench (2026) is an expert-annotated benchmark for multimodal dental triage: each outpatient case pairs a chief complaint with a panoramic radiograph, and a model predicts the full set of referral labels zero-shot. This board is the percentage of cases whose full set of coarse-grained domains is predicted exactly; higher is better.
Metric: Exact match (from 0-1, times 100): share of cases whose full set of 8 coarse-grained domains is predicted exactly, on the 246 Dental-TriageBench outpatient cases (chief complaint plus panoramic radiograph), zero-shot multi-label referral prediction; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 19 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 22.4 | #119 |
| 2 | Gemini 3 Flash (Preview) | 22 | #152 |
| 3 | GPT-5.2 | 21.5 | #185 |
| 4 | Kimi K2.5 | 18.7 | #266 |
| 5 | Grok 4 | 18.7 | #319 |
| 6 | Qwen-VL-Max-Latest (Dental-TriageBench checkpoint unspecified) | 18.3 | |
| 7 | Ministral-3-14B-Instruct-2512 | 16.3 | #1639 |
| 8 | MiMo-VL-7B-RL | 14.6 | #1155 |
| 9 | Claude Sonnet 4.5 | 13 | #261 |
| 10 | MedGemma-27B-IT | 13 | #1350 |
| 11 | Qwen 3 VL 8B Instruct | 10.6 | #1029 |
| 12 | Qwen 2.5 VL 7B Instruct | 10.2 | #1902 |
| 13 | Lingshu-7B | 9.4 | #1635 |
| 14 | GLM-4.1V-9B (Thinking) | 6.9 | |
| 15 | HuatuoGPT-Vision-7B | 6.1 | #2162 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=dental-triagebench-coarse-grained-exact-match · How It Works · Data refreshed daily, snapshot 2026-11-21.