Long-video understanding on VideoMME (average)
81.3AccuracyGemini 1.5 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 1.5 ProModel=Gemini 1.5 Pro2026.06 | 81.3 | |
| Claude 3.5 OpusModel=Claude 3.5 Opus2026.06 | 77.6 | |
| GPT-4oModel=GPT-4o2026.06 | 77.2 | |
| LLaVA-VideoParameters=72B2026.06 | 76.9 | |
| VideoRAGParameters=72B2026.06 | 75.7 | |
| InternVL-2.5Parameters=78B2026.06 | 74 | |
| AriaParameters=28B2026.06 | 72.1 | |
| HPPModel=Ours HPP2026.06 | 71.8 | |
| Dr. VideoModel=Dr. Video2026.06 | 71.7 | |
| LLaVA-VideoParameters=7B2026.06 | 69.7 | |
| LLaVA-OneVisionParameters=72B2026.06 | 69.6 | |
| VideoLLaMA-2Parameters=72B2026.06 | 64.7 | |
| LongVAParameters=7B2026.06 | 54.3 | |
| VideoChat2Parameters=7B2026.06 | 43.8 | |
| ShareGPT4VideoParameters=8B2026.06 | 43.6 |