Video Understanding on VideoEvalPro (test)
0.784AccuracyGemini 2.5 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 2.5 ProAccess Type=API call only2026.01 | 0.784 | |
| Gemini 2.5 ProCategory=API call only2026.03 | 0.784 | |
| Gemini 3 ProAccess Type=API call only2026.01 | 0.78 | |
| Gemini 3 ProCategory=API call only2026.03 | 0.78 | |
| Gemini 2.5 FlashAccess Type=API call only2026.01 | 0.696 | |
| Gemini 2.5 FlashCategory=API call only2026.03 | 0.696 | |
| GPT-5Access Type=API call only2026.01 | 0.688 | |
| GPT-5Category=API call only2026.03 | 0.688 | |
| Molmo2-8BAccess Type=Molmo2 family2026.01 | 0.604 | |
| Molmo2-8BCategory=Fully open2026.03 | 0.604 | |
| GPT-5 miniAccess Type=API call only2026.01 | 0.601 | |
| GPT-5 miniCategory=API call only2026.03 | 0.601 | |
| Molmo2-4BAccess Type=Molmo2 family2026.01 | 0.599 | |
| Molmo2-4BCategory=Fully open2026.03 | 0.599 | |
| Eagle2.5-8BAccess Type=Open weights only2026.01 | 0.586 | |
| Eagle2.5-8BCategory=Open weights2026.03 | 0.586 | |
| MolmoPoint-8BCategory=MolmoPoint2026.03 | 0.58 | |
| Molmo2-O-7BAccess Type=Molmo2 family2026.01 | 0.551 | |
| Molmo2-O-7BCategory=Fully open2026.03 | 0.551 | |
| Keye-VL-1.5-8BAccess Type=Open weights only2026.01 | 0.549 | |
| MiniCPM-V-4.5-8BAccess Type=Open weights only2026.01 | 0.549 | |
| Keye-VL-1.5-8BCategory=Open weights2026.03 | 0.549 | |
| MiniCPM-V-4.5-8BCategory=Open weights2026.03 | 0.549 | |
| VideoChat-Flash-7BAccess Type=Open models2026.01 | 0.512 | |
| VideoChat-Flash-7BCategory=Fully open2026.03 | 0.512 | |
| GLM-4.1V-9BAccess Type=Open weights only2026.01 | 0.511 | |
| GLM-4.1V-9BCategory=Open weights2026.03 | 0.511 | |
| Claude Sonnet 4.5Access Type=API call only2026.01 | 0.505 | |
| Claude Sonnet 4.5Category=API call only2026.03 | 0.505 | |
| Qwen3-VL-8BAccess Type=Open weights only2026.01 | 0.503 | |
| Qwen3-VL-8BCategory=Open weights2026.03 | 0.503 | |
| Qwen3-VL-4BAccess Type=Open weights only2026.01 | 0.498 | |
| Qwen3-VL-4BModel=Qwen3-VL-4B, Pruning Strength=None2026.03 | 0.498 | |
| Qwen3-VL-4BCategory=Open weights2026.03 | 0.498 | |
| InternVL3.5-8BAccess Type=Open weights only2026.01 | 0.481 | |
| InternVL3.5-8BModel=InternVL3.5-8B, Pruning Strength=None2026.03 | 0.481 | |
| InternVL3.5-8BCategory=Open weights2026.03 | 0.481 | |
| LLaVA-Video-7BAccess Type=Open models2026.01 | 0.478 | |
| LLaVA-Video-7BCategory=Fully open2026.03 | 0.478 | |
| STTSModel=STTS, Pruning Strength=30%2026.03 | 0.477 | |
| BaselineModel=Baseline, Pruning Strength=0%2026.03 | 0.476 | |
| PLM-8BAccess Type=Open models2026.01 | 0.472 | |
| PLM-8BModel=PLM-8B, Pruning Strength=None2026.03 | 0.472 | |
| STTSModel=STTS, Pruning Strength=40%2026.03 | 0.472 | |
| PLM-8BCategory=Fully open2026.03 | 0.472 | |
| InternVL3.5-4BAccess Type=Open weights only2026.01 | 0.465 | |
| InternVL3.5-4BCategory=Open weights2026.03 | 0.465 | |
| PLM-3BAccess Type=Open models2026.01 | 0.462 | |
| PLM-3BCategory=Fully open2026.03 | 0.462 | |
| STTSModel=STTS, Pruning Strength=50%2026.03 | 0.46 |