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

Multi-image Understanding on MuirBench Multi-image Understanding

62.3Accuracy

GPT-4V

Updated 1mo ago

Evaluation Results

Method	Links
GPT-4V 2025.12		62.3
Qwen2.5-VL-7B 2025.12		58.2
InternVL-2.5-8B 2025.12		51.2
SDAR-VL-8B-Inst 2025.12		50.2
InternVL-2-8B 2025.12		48.5
LLaDA-V-8B 2025.12		48.3
Qwen2.5-VL-3B 2025.12		46.5
InternVL-2.5-4B 2025.12		45.1
SDAR-VL-4B-Inst 2025.12		44.8
LLaVA-OV-7B 2025.12		40.5
InternVL-2-4B 2025.12		40.3
Qwen2-VL-7B 2025.12		39.9
GPT-4o 2025.12		0.68
MAmmoTH-VL 2025.12		0.551
Dream-VL 2025.12		0.512
LLaDA-V 2025.12		0.483
LLaVA-OV 2025.12		0.418