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

General Multimodal Evaluation on MMVet turbo

69.7Overall Score

D2Djus

Updated 1mo ago

Evaluation Results

Method	Links
D2Djus 2025.07		69.7
GPT-4o 2025.07		69.1
D2Iloc 2025.07		67.9
GPT-4V 2025.07		67.5
R1-Onevision-7B 2025.07		67.5
Qwen2.5-VL-7B* 2025.07		67.1
D2Ijus 2025.07		66.9
D2Dpar 2025.07		66.4
D2Dloc 2025.07		65.3
D2Ipar 2025.07		65.3
InternVL2.5-8B 2025.07		62.8
Qwen2-VL-7B 2025.07		62
Qwen2.5-VL-7B w/ GRPO 2025.07		60.8
LLaVA-CoT-11B 2025.07		60.3
Qwen2.5-VL-7B w/ GRPO† 2025.07		58.6
InternVL2-8B 2025.07		54.2