Video-MME
Multimodal · 2024-05-31
Video-MME evaluates multimodal language models on multiple-choice video question answering across diverse domains, durations, and modalities. Rows in this dataset use the overall no-subtitles metric unless a source note states otherwise.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 3 Pro Preview | 87.7 |
| GPT-5.2 | 85.8 |
| Gemini 2.5 Pro | 85.1 |
| GPT-5 | 84.7 |
| Qwen3.6 Plus (2026-04-02) | 84.2 |
| Qwen3.5 122B A10B | 83.9 |
| Qwen3.5 397B A17B | 83.7 |
| Kimi K2.5 | 83.2 |
| Qwen3.5-27B | 82.8 |
| Qwen3.5 35B A3B | 82.5 |
| Qwen3.6 35B A3B | 82.5 |
| Qwen3.5-Omni-Plus | 81.9 |
| Opus 4.5 | 81.4 |
| Qwen3.5 Plus (2026-02-15) | 81.0 |
| Gemini 2.5 Flash | 79.6 |
| Qwen3-VL-235B-A22B-Instruct | 79.2 |
| Qwen3 VL 235B A22B Thinking | 79.0 |
| GPT-5 Mini | 78.9 |
| Qwen3.5-9B | 78.4 |
| Qwen3 VL 32B Thinking | 77.3 |
| Qwen3.5-Omni-Flash | 77.0 |
| Qwen3.5-4B | 76.9 |
| Qwen3 VL 32B Instruct | 76.6 |
| GPT-5.6 Luna | 75.7 |
| Opus 4.1 | 75.6 |
| Sonnet 4.5 | 75.3 |
| GLM-4.5V | 74.6 |
| Qwen3 VL 30B A3B Instruct | 74.5 |
| Gemini 3.5 Flash-Lite | 74.0 |
| Qwen2.5-VL-72B-Instruct | 73.3 |
| Qwen3 VL 30B A3B Thinking | 73.3 |
| InternVL3.5-241B-A28B | 72.9 |
| Gemini 2.5 Flash-Lite | 72.7 |
| Gemini 2.0 Flash | 72.4 |
| Sonnet 4 | 72.3 |
| InternVL2.5-78B | 72.1 |
| Qwen3 VL 8B Thinking | 71.8 |
| Qwen3-VL-8B-Instruct | 71.4 |
| Qwen2 VL 72B Instruct | 71.2 |
| Qwen2.5 VL 32B Instruct | 70.5 |