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

General Visual Reasoning on M3CoT

74.2Accuracy

GPT4o

Updated 1mo ago

Evaluation Results

Method	Links
GPT4o 2026.02		74.2
ViGoRL 2026.02		68.72
SAYO-Qwen-8B 2026.02		68.46
NoisyRollout-7B 2026.02		67.9
Semantic-back-7B 2026.02		67.77
Qwen3-VL-4B 2026.02		65.1
Qwen3-VL-8B 2026.02		64.71
SAYO-Qwen-4B 2026.02		64.32
OpenVLThinker-7B 2026.02		63.29
R1-Onevision-7B 2026.02		61.73
SAYO-InternVL-8B 2026.02		57.2
Qwen3-VL-30B-A3B 2026.02		56.86
InternVL3.5-30B-A3B 2026.02		55.65
InternVL3.5-8B 2026.02		54.7
Kimi-VL-16B 2026.02		54.01
InternVL3.5-38B 2026.02		53.75
InternVL3.5-14B 2026.02		51.86