Multi-modal Video Evaluation on Video-MME
75AccuracyGemini-1.5-pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-1.5-proCategory=Proprietary Models2025.12 | 75 | |
| GPT-4oCategory=Proprietary Models2025.12 | 71.9 | |
| Qwen3-VL-4BCategory=Streamo Framework, Model Size=4B2025.12 | 69.3 | |
| FrameOracleBackbone=Qwen3-VL-8B, Frames=128→27.82025.10 | 69.1 | |
| Streamo-4BCategory=Streamo Framework, Model Size=4B, Base Model=Qwen3-VL-4B2025.12 | 68.7 | |
| Streamo-7BCategory=Streamo Framework, Model Size=7B, Base Model=Qwen2.5-VL-7B2025.12 | 67.9 | |
| FrameOracleBackbone=Qwen3-VL-8B, Frames=32→20.92025.10 | 67.3 | |
| Qwen3-VL-8BFrames=322025.10 | 66.9 | |
| Video-XL-2-8BFrames=∼10k2025.10 | 66.6 | |
| MACDBackbone=Qwen2.5-VL-7B2026.02 | 66 | |
| StreamingVLM-7BCategory=Open-source Online Models, Model Size=7B2025.12 | 65.1 | |
| Qwen2.5-VL-7BCategory=Streamo Framework, Model Size=7B2025.12 | 65.1 | |
| MACDBackbone=Qwen3-VL-2B2026.02 | 64 | |
| BaselineBackbone=Qwen2.5-VL-7B2026.02 | 64 | |
| Video-XL-7BFrames=2562025.10 | 64 | |
| MACDBackbone=Qwen2.5-VL-3B2026.02 | 62 | |
| Streamo-3BCategory=Streamo Framework, Model Size=3B, Base Model=Qwen2.5-VL-3B2025.12 | 61.8 | |
| FrameOracleBackbone=VideoLLaMA3-7B, Frames=64→13.92025.10 | 61.8 | |
| FrameOracleBackbone=LLaVA-Video-7B, Frames=64→13.92025.10 | 61.6 | |
| Qwen2.5-VL-3BCategory=Streamo Framework, Model Size=3B2025.12 | 61.5 | |
| FrameOracleBackbone=VideoLLaMA3-7B, Frames=16→10.42025.10 | 61.4 | |
| Flash-VStream-7BCategory=Open-source Online Models, Model Size=7B2025.12 | 61.2 | |
| VideoLLaMA3-7BFrames=162025.10 | 61.2 | |
| Streamo-2BCategory=Streamo Framework, Model Size=2B, Base Model=InternVL3-2B2025.12 | 60.1 | |
| VCDBackbone=Qwen2-VL-7B2026.02 | 60 | |
| LLaVA-Video-7BFrames=162025.10 | 59.8 | |
| FrameOracleBackbone=Qwen2.5-VL-3B, Frames=128→27.82025.10 | 59.7 | |
| FrameOracleBackbone=LLaVA-Video-7B, Frames=16→10.42025.10 | 59.6 | |
| MACDBackbone=Qwen2-VL-7B2026.02 | 59 | |
| InternVL3-2BCategory=Streamo Framework, Model Size=2B2025.12 | 58.9 | |
| FrameOracleBackbone=Qwen2.5-VL-3B, Frames=32→20.92025.10 | 58.5 | |
| Qwen2.5-VL-3BFrames=322025.10 | 58.4 | |
| FrameOracleBackbone=LLaVA-OneVision-7B, Frames=64→13.92025.10 | 58.1 | |
| Dispider-7BCategory=Open-source Online Models, Model Size=7B2025.12 | 57.2 | |
| LLaVA-OneVision-7BFrames=162025.10 | 56.1 | |
| SIDBackbone=Qwen3-VL-2B2026.02 | 56 | |
| SIDBackbone=Qwen2-VL-7B2026.02 | 56 | |
| FrameOracleBackbone=LLaVA-OneVision-7B, Frames=16→10.42025.10 | 56 | |
| BaselineBackbone=Qwen2.5-VL-3B2026.02 | 54 | |
| VCDBackbone=Qwen2.5-VL-7B2026.02 | 54 | |
| SIDBackbone=Qwen2.5-VL-7B2026.02 | 52 | |
| SIDBackbone=Qwen2.5-VL-3B2026.02 | 51 | |
| VCDBackbone=Qwen2.5-VL-3B2026.02 | 51 | |
| BaselineBackbone=Qwen2-VL-7B2026.02 | 51 | |
| MACDBackbone=InternVL3-8B2026.02 | 49 | |
| BaselineBackbone=InternVL3-8B2026.02 | 48 | |
| VideoLLaMA2-7BFrames=162025.10 | 47.9 | |
| BaselineBackbone=Qwen3-VL-2B2026.02 | 46 | |
| MACDBackbone=Qwen2-VL-2B2026.02 | 46 | |
| SIDBackbone=InternVL3-8B2026.02 | 46 | |
| BaselineBackbone=Qwen2-VL-2B2026.02 | 45 | |
| VCDBackbone=Qwen3-VL-2B2026.02 | 44 | |
| VCDBackbone=Qwen2-VL-2B2026.02 | 44 | |
| VCDBackbone=InternVL3-8B2026.02 | 44 | |
| SIDBackbone=Qwen2-VL-2B2026.02 | 43 | |
| VideoChat2-7BFrames=162025.10 | 39.5 | |
| VideoLLM-online-8BCategory=Open-source Online Models, Model Size=8B2025.12 | 26.9 |