Long Video Question Answering on Video-MME
73.2AccuracyToolMerge
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ToolMergeAnswerer model=GPT-4o, Frame budget (K)=322026.05 | 73.2 | — | — | — | — | — | — | — | — | |
| WFS (CVPR 2026)Answerer model=GPT-4o, Frame budget (K)=322026.05 | 72.6 | — | — | — | — | — | — | — | — | |
| AKS (CVPR 2025)Answerer model=GPT-4o, Frame budget (K)=322026.05 | 72.5 | — | — | — | — | — | — | — | — | |
| BOLT (CVPR 2025)Answerer model=GPT-4o, Frame budget (K)=322026.05 | 71.7 | — | — | — | — | — | — | — | — | |
| ToolMergeAnswerer model=GPT-4o, Frame budget (K)=82026.05 | 71 | — | — | — | — | — | — | — | — | |
| UniformAnswerer model=GPT-4o, Frame budget (K)=322026.05 | 70.6 | — | — | — | — | — | — | — | — | |
| ToolMergeAnswerer model=Qwen3-VL, Frame budget (K)=322026.05 | 70.6 | — | — | — | — | — | — | — | — | |
| SigLIP-QAnswerer model=GPT-4o, Frame budget (K)=322026.05 | 69.2 | — | — | — | — | — | — | — | — | |
| SigLIP-QAnswerer model=Qwen3-VL, Frame budget (K)=322026.05 | 68 | — | — | — | — | — | — | — | — | |
| WFS (CVPR 2026)Answerer model=Qwen3-VL, Frame budget (K)=322026.05 | 67.5 | — | — | — | — | — | — | — | — | |
| BOLT (CVPR 2025)Answerer model=Qwen3-VL, Frame budget (K)=322026.05 | 67.4 | — | — | — | — | — | — | — | — | |
| UniformAnswerer model=Qwen3-VL, Frame budget (K)=322026.05 | 67.3 | — | — | — | — | — | — | — | — | |
| AKS (CVPR 2025)Answerer model=Qwen3-VL, Frame budget (K)=322026.05 | 67.3 | — | — | — | — | — | — | — | — | |
| LIF (NeurIPS 2025)Answerer model=GPT-4o, Frame budget (K)=322026.05 | 67.2 | — | — | — | — | — | — | — | — | |
| BOLT (CVPR 2025)Answerer model=GPT-4o, Frame budget (K)=82026.05 | 67 | — | — | — | — | — | — | — | — | |
| WFS (CVPR 2026)Answerer model=GPT-4o, Frame budget (K)=82026.05 | 66.6 | — | — | — | — | — | — | — | — | |
| AKS (CVPR 2025)Answerer model=GPT-4o, Frame budget (K)=82026.05 | 64.7 | — | — | — | — | — | — | — | — | |
| ToolMergeAnswerer model=Qwen3-VL, Frame budget (K)=82026.05 | 64.6 | — | — | — | — | — | — | — | — | |
| SigLIP-QAnswerer model=GPT-4o, Frame budget (K)=82026.05 | 64.2 | — | — | — | — | — | — | — | — | |
| UniformAnswerer model=GPT-4o, Frame budget (K)=82026.05 | 64 | — | — | — | — | — | — | — | — | |
| LIF (NeurIPS 2025)Answerer model=GPT-4o, Frame budget (K)=82026.05 | 63.6 | — | — | — | — | — | — | — | — | |
| WFS (CVPR 2026)Answerer model=Qwen3-VL, Frame budget (K)=82026.05 | 63.3 | — | — | — | — | — | — | — | — | |
| LIF (NeurIPS 2025)Answerer model=Qwen3-VL, Frame budget (K)=322026.05 | 62.8 | — | — | — | — | — | — | — | — | |
| BOLT (CVPR 2025)Answerer model=Qwen3-VL, Frame budget (K)=82026.05 | 62.4 | — | — | — | — | — | — | — | — | |
| SigLIP-QAnswerer model=Qwen3-VL, Frame budget (K)=82026.05 | 58.5 | — | — | — | — | — | — | — | — | |
| AKS (CVPR 2025)Answerer model=Qwen3-VL, Frame budget (K)=82026.05 | 58.5 | — | — | — | — | — | — | — | — | |
| UniformAnswerer model=Qwen3-VL, Frame budget (K)=82026.05 | 57.9 | — | — | — | — | — | — | — | — | |
| LIF (NeurIPS 2025)Answerer model=Qwen3-VL, Frame budget (K)=82026.05 | 57 | — | — | — | — | — | — | — | — | |
| Blind Text (0 frames)Answerer model=Qwen3-VL, Frame budget (K)=02026.05 | 44.5 | — | — | — | — | — | — | — | — | |
| Blind Text (0 frames)Answerer model=GPT-4o, Frame budget (K)=02026.05 | 38.7 | — | — | — | — | — | — | — | — | |
| DrVideoModel Type=Agent Based Model2025.06 | — | — | — | — | — | — | — | 51.7 | 71.7 | |
| Gemini 1.5 ProModel Type=Closed-Source Model2025.06 | — | 81.7 | 84.5 | 74.3 | 81 | 67.4 | 77.4 | 75 | 81.3 | |
| GPT-4o(0513)Model Type=Closed-Source Model2025.06 | — | 80 | 82.8 | 70.3 | 76.6 | 65.3 | 72.1 | 71.9 | 77.2 | |
| InternVideo-2.5-8BModel Type=Open-Source 7B-8B Model2025.06 | — | — | — | — | — | — | — | 65.1 | — | |
| InternVL-2.5-78BModel Type=Open-Source 72B-78B Model2025.06 | — | 82.8 | 83.2 | 70.9 | 74.1 | 62.6 | 64.8 | 72.1 | 74 | |
| InternVL-2.5-8BModel Type=Open-Source 7B-8B Model2025.06 | — | — | — | — | — | — | — | 64.2 | 66.9 | |
| InternVL-3-78BModel Type=Open-Source 72B-78B Model2025.06 | — | — | — | — | — | — | — | 72.7 | 75.7 | |
| InternVL-3-8BModel Type=Open-Source 7B-8B Model2025.06 | — | — | — | — | — | — | — | 66.3 | 68.9 | |
| LLaVA-OneVision-72BModel Type=Open-Source 72B-78B Model2025.06 | — | 76.7 | 79.3 | 62.2 | 66.9 | 60 | 62.4 | 66.3 | 69.6 | |
| LLAVA-Video-72BModel Type=Open-Source 72B-78B Model2025.06 | — | 81.4 | 82.8 | 68.9 | 75.6 | 61.5 | 72.5 | 70.6 | 76.9 | |
| LLaVA-Video-7BModel Type=Open-Source 7B-8B Model2025.06 | — | — | — | — | — | — | — | 63.3 | 69.7 | |
| LongVU-7BModel Type=Open-Source 7B-8B Model2025.06 | — | — | — | — | — | — | 59.5 | — | 60.6 | |
| Qwen2.5-VL-72BModel Type=Open-Source 72B-78B Model2025.06 | — | 80.1 | 82.2 | 71.3 | 76.8 | 62.2 | 74.3 | 71.2 | 77.8 | |
| Qwen2.5-VL-7BModel Type=Open-Source 7B-8B Model2025.06 | — | — | — | — | — | — | — | 65.1 | 71.6 | |
| ShareGPT4Video-8BModel Type=Open-Source 7B-8B Model2025.06 | — | 48.3 | 53.6 | 36.3 | 39.3 | 35 | 37.9 | 39.9 | 43.6 | |
| T*Model Type=Agent Based Model2025.06 | — | 61 | — | 66.6 | — | 77.5 | — | 68.3 | — | |
| VideoChat-A1Backbone=Qwen2.5-VL-7B2025.06 | — | 78 | 81.2 | 73.1 | 77.4 | 64.3 | 70.9 | 71.8 | 76.5 | |
| VideoChat-A1Backbone=InternVL2.5-8B2025.06 | — | 80.8 | 81.4 | 72.3 | 76.8 | 63.7 | 72.6 | 72.3 | 77 | |
| VideoChat-A1Backbone=InternVideo2.5-8B2025.06 | — | 81.4 | 82.4 | 72.8 | 76.7 | 65 | 71.2 | 72.9 | 76.8 | |
| VideoChat2-7BModel Type=Open-Source 7B-8B Model2025.06 | — | 48.3 | 52.8 | 37 | 39.4 | 33.2 | 39.2 | 39.5 | 43.8 | |
| VideoLLaMA-2-72BModel Type=Open-Source 72B-78B Model2025.06 | — | 69.8 | 72 | 59.9 | 63 | 57.6 | 59 | 62.4 | 64.7 | |
| VideoMindModel Type=Agent Based Model2025.06 | — | — | — | — | — | 49.2 | — | 58.2 | — | |
| VideoRAG-72BModel Type=Agent Based Model2025.06 | — | 81.1 | — | 72.9 | — | 73.1 | — | 75.7 | — |