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

Video Reasoning on VideoEspresso

58.7Accuracy

STORM

Updated 8d ago

Evaluation Results

Method	Links
STORM 2026.05		58.7
Video-Thinker-7B 2026.05		56.4
LongVT-RFT 2026.05		52.7
Qwen2.5-VL-7B-SFT 2026.05		49.6
Qwen2.5-VL-7B-Instruct(CoT) 2026.05		45
Gemini-1.5-Pro 2026.05		44.2