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.
4rows
scoreprimary metric
2026-05-06sampled
Metadata
Metrics
Score, Normalized Score
| Rank | Subject | Score | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | Kimi K2.5 | 0.80 | MoonshotAI: Kimi K2.5 moonshotai-kimi-k2.5 | Self-reported | 2026-05-06 |
| 2 | Qwen3.5-122B-A10B | 0.75 |  Qwen3.5-122B-A10B qwen-qwen3.5-122b-a10b | Self-reported | 2026-05-06 |
| 3 | Qwen3.5-27B | 0.73 | Qwen3.5-27B qwen-qwen3.5-27b | Self-reported | 2026-05-06 |
| 4 | Qwen3.5-35B-A3B | 0.72 | Qwen3.5-35B-A3B qwen-qwen3.5-35b-a3b | Self-reported | 2026-05-06 |
No matching rows.