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

General Visual Reasoning on MME-RealWorld-Lite

73.06Accuracy

GPT4o

Updated 24d ago

Evaluation Results

Method	Links
GPT4o 2026.02		73.06
S1-VL-32B-RL 2026.04		67.1
SAYO-Qwen-8B 2026.02		62.85
S1-VL-32B-SFT 2026.04		60.8
Gemini-3-Flash 2026.06		60.34
Skywork-R1V4-30B 2026.04		59.3
PixelEyes 2026.06		59.25
Gemini 2.5 Pro 2026.04		58.3
SAYO-Qwen-4B 2026.02		57.63
GPT-5 2026.04		57.22
Qwen3-VL-4B 2026.02		56.8
Qwen3-VL-235B-A22B-Thinking 2026.04		56.5
Intern-S1 (235B+6B) 2026.04		56.48
Qwen3-VL-8B 2026.02		56.23
ViGoRL 2026.02		55.34
Thyme-VL (7B) 2026.04		55.2
InternVL3.5-38B 2026.02		54.56
PixelEyes 2026.06		54.51
Pixel-Reasoner 2026.06		54.32
DeepEyes 2026.06		53.53
NoisyRollout-7B 2026.02		51.69
Gemini 2.5 Flash 2026.04		50.2
Thyme 2026.06		50.13
SAYO-InternVL-8B 2026.02		50.03
Qwen-3-VL 2026.06		49.04
CVSearch 2026.05		48.8
InternVL3-8B 2026.04		48.6
InternVL3.5-14B 2026.02		48.57
OpenVLThinker-7B 2026.02		48.31
Qwen3-VL-32B-Thinking 2026.04		47.94
CVSearch 2026.05		46.7
Semantic-back-7B 2026.02		46.43
Qwen2.5-VL-32B 2026.04		46.2
CVSearch 2026.05		46.1
InternVL3.5-8B 2026.02		45.34
InternVL2.5-8B 2026.05		44.9
R1-Onevision-7B 2026.02		44.87
Qwen-3-VL 2026.06		44.55
Qwen-2.5-VL 2026.06		44.37
Qwen2.5-VL-7B 2026.04		44.1
Qwen3-VL-30B-A3B 2026.02		43.72
LLaVA-OV-7B 2026.05		43.7
Qwen2.5-VL-7B 2026.05		42.3
Mini-o3 2026.06		42.26
Kimi-VL-16B 2026.02		42
Intern-S1-mini (8B) 2026.04		40.02
InternVL3.5-30B-A3B 2026.02		38.61