Dental Visual Question Answering on MMMOral-OPG-Bench
60AccuracyDentalGPT
Evaluation results
| Method | Links | |
|---|---|---|
| DentalGPTComplex reasoning mode=true2025.12 | 60 | |
| LLaMA-4-MaverickComplex reasoning mode=false2025.12 | 51.4 | |
| Claude-Sonnet-4.5-ThinkingComplex reasoning mode=true2025.12 | 50.3 | |
| GPT-5Complex reasoning mode=true2025.12 | 47.7 | |
| GPT-4.1Complex reasoning mode=false2025.12 | 47.2 | |
| Grok-4.1-FastComplex reasoning mode=false2025.12 | 47.1 | |
| Claude-Sonnet-4.5Complex reasoning mode=false2025.12 | 47 | |
| GLM-4.5vComplex reasoning mode=true2025.12 | 45.7 | |
| Gemini-2.5-Pro-ThinkingComplex reasoning mode=true2025.12 | 45.7 | |
| Ernie-4.5-VL-424B-A47BComplex reasoning mode=true2025.12 | 45 | |
| Gemma-3-27B-itComplex reasoning mode=false2025.12 | 42.2 | |
| Mistral-Large-2512Complex reasoning mode=false2025.12 | 41.9 | |
| Qwen3-VL-235B-A22B-ThinkingComplex reasoning mode=true2025.12 | 40.6 | |
| Qwen3-VL-235B-A22B-InstructComplex reasoning mode=false2025.12 | 40.3 | |
| Deepseek-VL2Complex reasoning mode=false2025.12 | 39.1 | |
| Phi-4-Multimodal-InstructComplex reasoning mode=false2025.12 | 38.5 | |
| Qwen2.5-VL-7B-InstructComplex reasoning mode=false2025.12 | 27 |