LLM Stats (MMVU): leaderboard
LLM Stats mirror of MMVU (Multimodal Multi-disciplinary Video Understanding), testing comprehension and reasoning over video content across disciplines.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around January 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-5.3 Flash | 80.5 | #65 |
| 2 | Kimi K2.5 | 80.4 | #144 |
| 3 | Qwen 3.5 122B A10B | 74.7 | #173 |
| 4 | Qwen 3.5 27B | 73.3 | #187 |
| 5 | Seed 1.8 | 73.1 | #139 |
| 6 | Qwen 3.5 35B A3B | 72.3 | #256 |
| 7 | Seed 2.0 Mini | 69 | #276 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-mmvu · How It Works · Data refreshed daily, snapshot 2026-10-20.