OpenVLM Video: leaderboard
OpenCompass OpenVLM video leaderboard aggregating video-language model results on Video-MME, MVBench, MMBench-Video, TempCompass, and MLVU.
Metric: Average Normalized Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 37 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL3-78B | 72.7 | #352 |
| 2 | InternVL3-38B | 71.9 | #405 |
| 3 | InternVL2.5-78B | 68.1 | #431 |
| 4 | Qwen 2 VL 72B | 67.7 | #460 |
| 5 | InternVL3-8B | 66.8 | #620 |
| 6 | Gemini 2.0 Flash | 64.6 | #342 |
| 7 | Qwen2.5-Omni-7B | 64.5 | #624 |
| 8 | Qwen 2.5 VL 7B | 62.8 | #643 |
| 9 | Qwen 2 VL 7B | 61.2 | #750 |
| 10 | Gemini 1.5 Flash | 57 | #618 |
| 11 | InternVL2-8B | 56.4 | #817 |
| 12 | MiniCPM-V-2.6 | 54.7 | #836 |
| 13 | Claude 3.5 Sonnet | 45.9 | #350 |
| 14 | Phi-4 Multimodal Instruct | 42.3 | #906 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-video · How It Works · Data refreshed daily, snapshot 2026-10-20.