Share your thoughts, 1 month free Claude Pro on usSee more

Dental Visual Question Answering on DentalBench Mixed

57.4Accuracy

Gemini-2.5-Pro-Thinking

Updated 1mo ago

Evaluation Results

Method	Links
Gemini-2.5-Pro-Thinking 2025.12		57.4
DentalGPT 2025.12		54.4
GPT-5 2025.12		54.3
LLaMA-4-Maverick 2025.12		53.9
Claude-Sonnet-4.5-Thinking 2025.12		53.9
Grok-4.1-Fast 2025.12		52.2
GPT-4.1 2025.12		51.7
Qwen3-VL-235B-A22B-Instruct 2025.12		51.6
Qwen3-VL-235B-A22B-Thinking 2025.12		51.6
Ernie-4.5-VL-424B-A47B 2025.12		51.4
GLM-4.5v 2025.12		51.4
Claude-Sonnet-4.5 2025.12		50.4
Mistral-Large-2512 2025.12		48.2
Qwen2.5-VL-7B-Instruct 2025.12		46.1
Phi-4-Multimodal-Instruct 2025.12		44.4
Gemma-3-27B-it 2025.12		43
Deepseek-VL2 2025.12		22.6