Video Reasoning and Long Video Understanding on Minerva
66.2AVG ScoreGemini-2.5-Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-2.5-ProCategory=Proprietary Models, Size=-2026.06 | 66.2 | |
| GPT-4oCategory=Proprietary Models, Size=-2026.06 | 45.5 | |
| OmniAgentCategory=Ours, Size=7B, incorporate audio signals=true2026.06 | 41.4 | |
| Qwen2.5-OmniCategory=Open-Source Non-Thinking Models, Size=7B, incorporate audio signals=true2026.06 | 33.4 | |
| Qwen2.5-VLCategory=Open-Source Non-Thinking Models, Size=7B2026.06 | 33 | |
| Video-CoMCategory=Open-Source Agentic Models, Size=7B2026.06 | 31.7 | |
| VideoRFTCategory=Open-Source Thinking Models, Size=7B2026.06 | 29.2 | |
| Video-R1Category=Open-Source Thinking Models, Size=7B2026.06 | 29.1 | |
| LongVTCategory=Open-Source Agentic Models, Size=7B2026.06 | 28.5 |