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

Social Reasoning on MotiveBench

94Accuracy

O3

Updated 1mo ago

Evaluation Results

Method	Links
O3 2026.03		94
GPT-4o 2026.03		93
GPT-5 2026.03		92.67
O3 2026.03		92
GPT-5 2026.03		90.89
GPT-4o 2026.03		89.78
Qwen3-32B 2026.03		89.22
SocialR1-8B 2026.03		88.89
LLaMa3.1-70B 2026.03		88.67
Distill-Llama-70B 2026.03		88.44
Qwen3-32B 2026.03		88
SocialR1-8B 2026.03		87.89
SocialR1-8B 2026.03		87.67
SocialR1-8B 2026.03		87.56
DeepSeek-R1 2026.03		86.55
SocialR1-8B 2026.03		85.67
SocialR1-4B 2026.03		85.33
SocialR1-4B 2026.03		85
LLaMa3.1-70B 2026.03		84.45
Qwen3-4B 2026.03		84.44
SocialR1-4B 2026.03		83.11
SocialR1-4B 2026.03		83.11
SocialR1-4B 2026.03		83
Qwen3-8B 2026.03		79.33
Qwen3-4B 2026.03		77.34
Qwen3-8B 2026.03		68.78