Long Video QA on Video-MME
84.3Average ScoreGemini 2.5 Pro
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemini 2.5 ProModel Category=Closed-source Large-sized MLLM2025.11 | 84.3 | — | — | |
| VideoChat-M1Model Category=Ours, Number of Parameters=37B2025.11 | 83.2 | 84.2 | 76.7 | |
| Qwen3-VL-235B-InstructModel Category=Open-source Large-sized MLLM2025.11 | 79.2 | — | — | |
| Qwen3-VL-235B-ThinkingModel Category=Open-source Large-sized MLLM2025.11 | 79 | — | — | |
| Seed 1.5VLModel Category=Closed-source Large-sized MLLM2025.11 | 77.9 | — | — | |
| VideoRAG-72BModel Category=Agent-based Methods2025.11 | 75.7 | 72.9 | 73.1 | |
| ReAgent-V-72BModel Category=Agent-based Methods2025.11 | 75.1 | 72.3 | 72.9 | |
| Gemini 1.5 ProModel Category=Closed-source Large-sized MLLM2025.11 | 75 | 74.3 | 67.4 | |
| Qwen2.5-VL-72BModel Category=Open-source Large-sized MLLM2025.11 | 73.3 | — | — | |
| InternVL-3.5-241BModel Category=Open-source Large-sized MLLM2025.11 | 72.9 | — | — | |
| VideoChat-A1Model Category=Agent-based Methods2025.11 | 72.9 | 72.8 | 65 | |
| InternVL-3-78BModel Category=Open-source Large-sized MLLM2025.11 | 72.7 | — | — | |
| Eagle-2.5-8BModel Category=Open-source Medium-sized MLLM2025.11 | 72.4 | — | — | |
| InternVL-2.5-78BModel Category=Open-source Large-sized MLLM2025.11 | 72.1 | 70.9 | 62.6 | |
| GPT-4oModel Category=Closed-source Large-sized MLLM2025.11 | 71.9 | 70.3 | 65.3 | |
| Qwen3-VL-8B-InstructModel Category=Open-source Medium-sized MLLM2025.11 | 71.4 | — | — | |
| Qwen2-VL-72BModel Category=Open-source Large-sized MLLM2025.11 | 71.2 | 71.3 | 62.2 | |
| InternVL-3.5-78BModel Category=Open-source Large-sized MLLM2025.11 | 70.9 | — | — | |
| LLAVA-Video-72BModel Category=Open-source Large-sized MLLM2025.11 | 70.6 | 68.9 | 61.5 | |
| Aria-28BModel Category=Open-source Large-sized MLLM2025.11 | 67.6 | 67 | 58.8 | |
| Oryx-34BModel Category=Open-source Large-sized MLLM2025.11 | 67.3 | 65.3 | 59.3 | |
| VideoChat-R1.5-7BModel Category=Open-source Medium-sized MLLM2025.11 | 67.1 | — | — | |
| VideoXL2-8BModel Category=Open-source Medium-sized MLLM2025.11 | 66.6 | — | — | |
| LLaVA-ov-72BModel Category=Open-source Large-sized MLLM2025.11 | 66.3 | 62.2 | 60 | |
| InternVL-3-8BModel Category=Open-source Medium-sized MLLM2025.11 | 66.3 | — | — | |
| InternVL-3.5-8BModel Category=Open-source Medium-sized MLLM2025.11 | 66 | — | — | |
| VideoChat-Flash-7BModel Category=Open-source Medium-sized MLLM2025.11 | 65.3 | — | 55.4 | |
| InternVideo2.5-7BModel Category=Open-source Medium-sized MLLM2025.11 | 65.1 | — | — | |
| Qwen2.5-VL-7BModel Category=Open-source Medium-sized MLLM2025.11 | 65.1 | — | — | |
| LongRL-7BModel Category=Open-source Medium-sized MLLM2025.11 | 65.1 | 63.2 | 55.2 | |
| InternVL-2.5-8BModel Category=Open-source Medium-sized MLLM2025.11 | 64.2 | — | — | |
| VITAL-7BModel Category=Agent-based Methods2025.11 | 64.1 | — | 54 | |
| Qwen2-VL-7BModel Category=Open-source Medium-sized MLLM2025.11 | 63.3 | — | — | |
| LLaVA-Video-7BModel Category=Open-source Medium-sized MLLM2025.11 | 63.3 | — | — | |
| VideoLLaMA2-72BModel Category=Open-source Large-sized MLLM2025.11 | 62.4 | 59.9 | 57.6 | |
| Video-R1-7BModel Category=Open-source Medium-sized MLLM2025.11 | 61.4 | — | — | |
| LongVILA-7BModel Category=Open-source Medium-sized MLLM2025.11 | 60.1 | 58.3 | 53 | |
| LongVA-7BModel Category=Open-source Medium-sized MLLM2025.11 | 52.6 | 50.4 | 46.2 | |
| DrVideoModel Category=Agent-based Methods2025.11 | 51.7 | — | — | |
| ShareGPT4Video-8BModel Category=Open-source Medium-sized MLLM2025.11 | 39.9 | 36.3 | 35 | |
| VideoChat2-7BModel Category=Open-source Medium-sized MLLM2025.11 | 39.5 | 37 | 33.2 | |
| DeepVideoDiscoveryModel Category=Agent-based Methods2025.11 | — | — | 67.3 |