VideoMMMU
multimodal
+
+
+
+
About
Video-MMMU is a massive multi-modal, multi-disciplinary video benchmark evaluating large multimodal models' knowledge acquisition capabilities from educational videos across diverse academic disciplines. Featuring 300 expert-level videos and 900 human-annotated questions, this comprehensive evaluation tests perception, comprehension, and adaptation abilities, measuring how effectively AI models learn from educational video content.
+
+
+
+
Evaluation Stats
Total Models4
Organizations2
Verified Results0
Self-Reported4
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers
Score Distribution
4 models
Top Score
84.6%
Average Score
78.2%
High Performers (80%+)
3Top Organizations
#1Google
1 model
83.6%
#2OpenAI
3 models
76.4%
+
+
+
+
Leaderboard
4 models ranked by performance on VideoMMMU
