Leaderboard - Open

Model Easy Med Hard Average
GPT-4o45.7%41.5%38.0%41.7%
Gemini-1.5-Pro30.3%30.5%24.1%28.3%
Claude-3.5-Sonnet37.8%34.6%34.3%35.6%
LLaVA-Video-7B7.8%9.0%8.5%8.4%
Qwen2VL-7B11.2%8.8%1.6%7.2%
VidDiff (ours)49.9%37.9%38.5%42.1%