Real-world Understanding on WildVision

80.6Win Rate

GPT-4o-20240513

Updated 11d ago

Evaluation Results

Method	Links
GPT-4o-20240513 2024.12		80.6
GPT-4V 2024.12		71.8
InternVL2.5-78B 2024.12		71.4
LLaVA-OneVision-7B w/ UnifiedReward-Think 2026.04		68.3
LLaVA-OneVision-7B w/ VL-MDR 2026.04		68.3
LLaVA-OneVision-7B w/ Skywork-VL 2026.04		68.2
LLaVA-OneVision-7B w/ IXC-2.5-Reward 2026.04		67.9
LLaVA-OneVision-7B w/ UnifiedReward 2026.04		67.8
LLaVA-OneVision-7B w/ LLaVA-Critic 2026.04		67.3
LLaVA-OneVision-7B w/ MM-RLHF-Reward 2026.04		67.1
InternVL2.5-38B 2024.12		66.4
InternVL2-Llama3-76B 2024.12		65.8
InternVL2.5-26B 2024.12		65.2
InternVL2-40B 2024.12		63.2
InternVL2-26B 2024.12		62.2
InternVL2.5-8B 2024.12		62
InternVL-Chat-V1.5 2024.12		56.6
LLaVA-OneVision-7B 2026.04		54.9
InternVL2-8B 2024.12		54.4
InternVL2.5-4B 2024.12		49.4
InternVL2.5-2B 2024.12		44.2
InternVL2-4B 2024.12		44.2
InternVL2.5-1B 2024.12		43.4
InternVL2-2B 2024.12		31.8
InternVL2-1B 2024.12		17.8