multimodal benchmark
MMVU
MMVU (Multimodal Multi-disciplinary Video Understanding) is a benchmark for evaluating multimodal models on video understanding tasks across multiple disciplines, testing comprehension and reasoning capabilities on video content.
Updated Aug 7, 2026
Published models4
Registry coverage4
MetricScore
EvidenceB
MMVU leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | MA | 80.4% | 100.0% | 4 | C | |
| 2 | AC | 74.7% | 66.7% | 4 | C | |
| 3 | AC | 73.3% | 33.3% | 4 | C | |
| 4 | AC | 72.3% | 0.0% | 4 | C |
Score distribution
Top published rows on the benchmark's original scale.
MMVU highlights
The top published results on this benchmark's own scale.
What is MMVU?
Definition and scoring fields from the benchmark registry.
MMVU (Multimodal Multi-disciplinary Video Understanding) is a benchmark for evaluating multimodal models on video understanding tasks across multiple disciplines, testing comprehension and reasoning capabilities on video content.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- MMVU
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- mmvu|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about MMVU.
Which model scores highest on MMVU?
Kimi K2.5 is currently ranked first with 80.4%.
What does MMVU measure?
MMVU (Multimodal Multi-disciplinary Video Understanding) is a benchmark for evaluating multimodal models on video understanding tasks across multiple disciplines, testing comprehension and reasoning capabilities on video content.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
4 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.