LLM Stats (VideoMMMU) — leaderboard
Metric: Score (%). Source: llm-stats.com. 26 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 87.6 |
| 2 | Gemini 3 Flash | 86.9 |
| 3 | Kimi K2.5 | 86.6 |
| 4 | GPT-5.2 | 85.9 |
| 5 | Qwen 3.7 Plus | 85.4 |
| 6 | Gemini 3.1 Flash Lite | 84.8 |
| 7 | GPT-5 | 84.6 |
| 8 | MiniMax-M3 | 84.6 |
| 9 | Qwen 3.6 27B | 84.4 |
| 10 | Qwen 3.6 Plus | 84 |
| 11 | Qwen 3.6 35B A3B | 83.7 |
| 12 | Gemini 2.5 Pro (Preview 06-05) | 83.6 |
| 13 | O3 | 83.3 |
| 14 | Qwen 3.5 27B | 82.3 |
| 15 | Qwen 3.5 122B A10B | 82 |
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-videommmu · How the rankings work · Data refreshed daily, snapshot 2026-07-22.