Video Understanding on Video-Native Challenges Video-Oasis
40.2Fine PerceptionGemini-2.5 Pro
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Gemini-2.5 ProModel Category=Proprietary LLMs2026.03 | 40.2 | 49.8 | 50.9 | 45.4 | 43 | 46.7 | |
| VideoToolModel Category=Agentic Methods, Reasoning Model=GPT-5-mini, Distilled evaluation setting=true2026.03 | 31.6 | 44.4 | 42.2 | 34 | 32.9 | 39.5 | |
| InternVL-3.5Model Category=Open Source Video-LLMs, Parameter Scale=8B2026.03 | 29.5 | 41.9 | 35.1 | 29.8 | 23.3 | 33.6 | |
| Qwen3-VLthink.Model Category=Open Source Video-LLMs, Parameter Scale=8B2026.03 | 29 | 41.6 | 37.7 | 27.7 | 23.2 | 34.6 | |
| VideoTreeModel Category=Agentic Methods, Reasoning Model=GPT-5-mini, Distilled evaluation setting=true2026.03 | 28.6 | 34 | 32.3 | 24.6 | 20.7 | 30.1 | |
| LongViLA-R1Model Category=Open Source Video-LLMs, Parameter Scale=7B2026.03 | 28.4 | 25.4 | 31.5 | 27.9 | 20.6 | 28.6 | |
| VideoAuto-R1Model Category=Open Source Video-LLMs, Parameter Scale=8B2026.03 | 27.5 | 44.3 | 39.5 | 31.1 | 28.9 | 36.8 | |
| Qwen3-VLinst.Model Category=Open Source Video-LLMs, Parameter Scale=8B2026.03 | 27 | 42.4 | 36.5 | 28 | 21.5 | 33.8 | |
| InternVL-3Model Category=Open Source Video-LLMs, Parameter Scale=8B2026.03 | 27 | 31.3 | 34.1 | 30.6 | 24.5 | 31.6 | |
| Eagle2.5Model Category=Open Source Video-LLMs, Parameter Scale=8B2026.03 | 26.9 | 31 | 39.7 | 33.2 | 22.7 | 34.5 | |
| GPT-4oModel Category=Proprietary LLMs2026.03 | 25.6 | 33.2 | 26.3 | 27.3 | 26.5 | 27.5 | |
| Video-R1Model Category=Open Source Video-LLMs, Parameter Scale=7B2026.03 | 24 | 24 | 29.1 | 27.3 | 18.4 | 26.3 | |
| Qwen2.5-VLModel Category=Open Source Video-LLMs, Parameter Scale=7B2026.03 | 23.3 | 28.7 | 32.3 | 28.6 | 21.2 | 29.2 |