Streaming Video Understanding on OVO-Bench
93.2Real-Time Visual Perception Avg.Human
Evaluation results
| Method | Links | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Human#Frames=–2026.04 | 93.2 | 94 | 92.6 | 94.8 | 92.7 | 91.1 | 94 | 92.6 | 93 | 91.4 | 92.3 | — | 92.77 | — | — | — | |
| SELECTSTREAM-Qwen3-VL-8B#Frames=1 fps2026.06 | 82.76 | — | — | — | — | — | — | — | — | — | 62.2 | — | 67.03 | 56.13 | — | 72.48 | |
| SIMPLESTREAM (Qwen3-VL-8B + 4f)#Frames=42026.04 | 81.4 | 94 | 85.3 | 82.8 | 65.7 | 77.2 | 83.2 | 51.9 | 58.1 | 52.1 | 54 | — | 67.7 | — | — | — | |
| Qwen3-VL-8B + 4f#Frames=1fps2026.06 | 81.4 | — | — | — | — | — | — | — | — | — | 54 | — | — | — | — | 67.7 | |
| SELECTSTREAM-Qwen2.5-VL-7B#Frames=1 fps2026.06 | 80.85 | — | — | — | — | — | — | — | — | — | 61.05 | — | 65.71 | 55.23 | — | 70.95 | |
| SIMPLESTREAM (Qwen3-VL-8B + 8f)#Frames=82026.04 | 79.9 | 94 | 84.4 | 80.2 | 64 | 75.3 | 81.5 | 53.2 | 60.8 | 50.5 | 54.9 | — | 67.37 | — | — | — | |
| SIMPLESTREAM (Qwen3-VL-8B + 2f)#Frames=22026.04 | 79.3 | 89.3 | 77.1 | 83.6 | 68.5 | 76.2 | 81 | 49.5 | 56.1 | 54.8 | 53.5 | — | 66.38 | — | — | — | |
| SIMPLESTREAM (Qwen2.5-VL-7B + 4f)#Frames=42026.04 | 78.4 | 94 | 72.5 | 80.2 | 68 | 76.2 | 79.3 | 54.5 | 60.8 | 40.3 | 51.9 | — | 65.13 | — | — | — | |
| Qwen2.5-VL-7B + 4f#Frames=1fps2026.06 | 78.4 | — | — | — | — | — | — | — | — | — | 51.9 | — | — | — | — | 65.13 | |
| SIMPLESTREAM (Qwen2.5-VL-7B + 8f)#Frames=82026.04 | 76.6 | 95.3 | 67.9 | 79.3 | 61.2 | 74.3 | 81.5 | 52.2 | 63.5 | 36.6 | 50.8 | — | 63.7 | — | — | — | |
| SIMPLESTREAM (Qwen2.5-VL-7B + 2f)#Frames=22026.04 | 75 | 88.6 | 67 | 81 | 64.6 | 69.3 | 79.3 | 49.2 | 56.8 | 42.5 | 49.5 | — | 62.22 | — | — | — | |
| Response-G1Architecture=Qwen3-VL-8B2026.05 | 73.6 | — | — | — | — | — | — | — | — | — | 52.1 | — | 61.3 | 58.2 | — | — | |
| R3-Streaming-7B|4B-ThinkingProcessing Mode=Streaming, Visual-token reduction ratio=96%↓, Backbone=7B|4B-Thinking2026.05 | 71.89 | — | — | — | — | — | — | — | — | — | 51.27 | — | 57.92 | — | 50.6 | — | |
| ThinkStream-7B#Frames=1 fps2026.06 | 69.12 | — | — | — | — | — | — | — | — | — | 60.68 | — | — | — | — | 64.9 | |
| HERMES-7B†#Frames=1 fps, Memory Budget=4K tokens2026.04 | 69 | 85.2 | 64.2 | 71.6 | 53.4 | 74.3 | 65.2 | 48.5 | 62.2 | 37.6 | 49.4 | — | 59.2 | — | — | — | |
| HERMES-7B#Frames=1 fps2026.06 | 69 | — | — | — | — | — | — | — | — | — | 49.4 | — | — | — | — | 59.2 | |
| Streamo-7B#Frames=2 fps2026.06 | 67.44 | — | — | — | — | — | — | — | — | — | 49.18 | — | 57.86 | 56.96 | — | 58.31 | |
| ThinkStream-3BModel Type=Online, Model Scale=3B2026.03 | 67.03 | 85.23 | 64.22 | 69.82 | 49.43 | 69.31 | 64.13 | 53.87 | 59.46 | 43.55 | 52.3 | 59.66 | — | — | — | — | |
| Response-G1Architecture=LLaVA-OV-1.5-8B2026.05 | 66.7 | — | — | — | — | — | — | — | — | — | 48.3 | — | 56.1 | 54.8 | — | — | |
| Streamo-7B#Frames=1 fps2026.04 | 66 | 79.2 | 57.8 | 75 | 49.4 | 64.4 | 70.1 | 54.6 | 52 | 31.7 | 46.1 | — | 56.05 | — | — | — | |
| Streamo-7BProcessing Mode=Streaming, Params=7B2026.05 | 65.98 | — | — | — | — | — | — | — | — | — | 46.1 | — | 55.61 | — | 54.77 | — | |
| Streamo-7B#Frames=1 fps2026.06 | 65.98 | — | — | — | — | — | — | — | — | — | 46.1 | — | 55.61 | 54.77 | — | 56.04 | |
| R3-Streaming-3B|7B-InstructProcessing Mode=Streaming, Visual-token reduction ratio=96%↓, Backbone=3B|7B-Instruct2026.05 | 65.12 | — | — | — | — | — | — | — | — | — | 44.64 | — | 54.07 | — | 52.44 | — | |
| Qwen2.5-VL-7BProcessing Mode=Offline, Params=7B2026.05 | 64.7 | — | — | — | — | — | — | — | — | — | 44.68 | — | 50.39 | — | 41.8 | — | |
| Qwen2.5-VL-32BModel Type=Offline, Model Scale=32B2026.03 | 64.4 | 77.18 | 58.72 | 68.1 | 50.56 | 74.26 | 57.61 | 58.59 | 62.84 | 29.57 | 50.33 | 57.37 | — | — | — | — | |
| LLaVA-OneVision-7BProcessing Mode=Offline, Params=7B2026.05 | 64.02 | — | — | — | — | — | — | — | — | — | 43.71 | — | 52.74 | — | 50.5 | — | |
| LLaVA-OneVision-7B#Frames=322026.04 | 64 | 66.4 | 57.8 | 73.3 | 53.4 | 71.3 | 62 | 54.2 | 55.4 | 21.5 | 43.7 | — | 53.85 | — | — | — | |
| LLaVA-OneVision-7B#Frames=322026.06 | 64 | — | — | — | — | — | — | — | — | — | 43.7 | — | 52.74 | 50.5 | — | 53.85 | |
| R3-Streaming-3B|4B-ThinkingProcessing Mode=Streaming, Visual-token reduction ratio=96%↓, Backbone=3B|4B-Thinking2026.05 | 63.58 | — | — | — | — | — | — | — | — | — | 50.23 | — | 56.55 | — | 55.83 | — | |
| LLaVA-Video-7BModel Type=Offline, Model Scale=7B2026.03 | 63.52 | 69.13 | 58.72 | 68.83 | 49.44 | 74.26 | 59.78 | 56.23 | 57.43 | 7.53 | 40.4 | 51.96 | — | — | — | — | |
| LLaVA-Video-7B#Frames=642026.04 | 63.5 | 69.1 | 58.7 | 68.8 | 49.4 | 74.3 | 59.8 | 56.2 | 57.4 | 7.5 | 40.4 | — | 51.95 | — | — | — | |
| LLaVA-Video-7B#Frames=642026.06 | 63.5 | — | — | — | — | — | — | — | — | — | 40.4 | — | 52.91 | 54.82 | — | 51.95 | |
| Qwen3-VL-4B-ThinkingProcessing Mode=Offline, Params=4B2026.05 | 62.93 | — | — | — | — | — | — | — | — | — | 51.56 | — | 57.74 | — | 58.74 | — | |
| Qwen2-VL-72BProcessing Mode=Offline, Params=72B2026.05 | 61.92 | — | — | — | — | — | — | — | — | — | 56.95 | — | 56.27 | — | 49.3 | — | |
| TimeChat-Online-7B#Frames=1 fps2026.04 | 61.9 | 75.2 | 46.8 | 70.7 | 47.8 | 69.3 | 61.4 | 55.9 | 59.5 | 9.7 | 41.7 | — | 51.8 | — | — | — | |
| TimeChat-Online-7B#Frames=1 fps2026.06 | 61.9 | — | — | — | — | — | — | — | — | — | 41.7 | — | — | — | — | 51.8 | |
| Streamo-3BModel Type=Online, Model Scale=3B2026.03 | 61.51 | 78.52 | 52.29 | 67.24 | 44.38 | 55.45 | 71.2 | 51.18 | 57.43 | 16.67 | 41.76 | 51.64 | — | — | — | — | |
| Streamo-3BProcessing Mode=Streaming, Params=3B2026.05 | 61.51 | — | — | — | — | — | — | — | — | — | 41.76 | — | 52.33 | — | 53.72 | — | |
| StreamAgent2026.05 | 61.3 | — | — | — | — | — | — | — | — | — | 41.7 | — | 49.4 | 45.4 | — | — | |
| StreamAgent-7BProcessing Mode=Streaming, Params=7B2026.05 | 61.3 | — | — | — | — | — | — | — | — | — | 41.7 | — | 49.4 | — | 45.4 | — | |
| StreamForest-7BModel Type=Online, Model Scale=7B2026.03 | 61.2 | 68.46 | 53.21 | 71.55 | 47.75 | 65.35 | 60.87 | 58.92 | 64.86 | 32.26 | 52.02 | 56.61 | — | — | — | — | |
| StreamForest-7B#Frames=1 fps2026.04 | 61.2 | 68.5 | 53.2 | 71.6 | 47.8 | 65.4 | 60.9 | 58.9 | 64.9 | 32.3 | 52 | — | 56.6 | — | — | — | |
| StreamForest-7B#Frames=1 fps2026.06 | 61.2 | — | — | — | — | — | — | — | — | — | 52 | — | — | — | — | 56.6 | |
| Qwen2-VL-7B#Frames=642026.04 | 60.7 | 69.1 | 53.2 | 63.8 | 50.6 | 66.3 | 60.9 | 44.4 | 66.9 | 34.4 | 48.6 | — | 54.62 | — | — | — | |
| Qwen2-VL-7B#Frames=642026.06 | 60.7 | — | — | — | — | — | — | — | — | — | 48.6 | — | — | — | — | 54.62 | |
| InternVL2-8B#Frames=162026.04 | 60.4 | 67.1 | 60.6 | 63.8 | 46.1 | 68.3 | 56.5 | 48.2 | 57.4 | 24.7 | 43.4 | — | 51.9 | — | — | — | |
| InternVL2-8B#Frames=162026.06 | 60.4 | — | — | — | — | — | — | — | — | — | 43.4 | — | 50.15 | 46.6 | — | 51.9 | |
| Qwen2.5-VL-3BModel Type=Offline, Model Scale=3B2026.03 | 60.03 | 76.51 | 44.03 | 67.24 | 42.13 | 68.31 | 61.96 | 50.5 | 53.38 | 22.04 | 41.98 | 51 | — | — | — | — | |
| Qwen2.5-VL-7BModel Type=Offline, Model Scale=7B2026.03 | 59.9 | 67.79 | 55.05 | 67.24 | 42.13 | 66.34 | 60.87 | 51.52 | 58.78 | 23.66 | 44.65 | 52.28 | — | — | — | — | |
| Qwen2.5-VL-7B#Frames=1 fps2026.04 | 59.9 | 67.8 | 55.1 | 67.2 | 42.1 | 66.3 | 60.9 | 51.5 | 58.8 | 23.7 | 44.7 | — | 52.28 | — | — | — | |
| Qwen2.5-VL-7B#Frames=1 fps2026.06 | 59.9 | — | — | — | — | — | — | — | — | — | 44.7 | — | — | — | — | 52.28 | |
| TimeChat-Online2026.05 | 58.6 | — | — | — | — | — | — | — | — | — | 42 | — | 45.6 | 36.4 | — | — | |
| TimeChat-Online-7BProcessing Mode=Streaming, Visual-token reduction ratio=85%↓, Params=7B2026.05 | 58.6 | — | — | — | — | — | — | — | — | — | 42 | — | 45.6 | — | 36.4 | — | |
| LongVU-7BModel Type=Offline, Model Scale=7B2026.03 | 57.61 | 53.69 | 53.21 | 62.93 | 47.75 | 68.32 | 59.78 | 40.74 | 59.46 | 4.84 | 35.01 | 46.31 | — | — | — | — | |
| LongVU-7BProcessing Mode=Offline, Params=7B2026.05 | 57.61 | — | — | — | — | — | — | — | — | — | 35.01 | — | 46.71 | — | 47.5 | — | |
| LongVU-7B#Frames=1 fps2026.04 | 57.4 | 55.7 | 49.5 | 59.5 | 48.3 | 68.3 | 63 | 43.1 | 66.2 | 9.1 | 39.5 | — | 48.45 | — | — | — | |
| LongVU-7B#Frames=1 fps2026.06 | 57.4 | — | — | — | — | — | — | — | — | — | 39.5 | — | — | — | — | 48.45 | |
| Qwen2-VL-7BModel Type=Offline, Model Scale=7B2026.03 | 55.98 | 60.4 | 50.46 | 56.03 | 47.19 | 66.34 | 55.43 | 47.81 | 35.48 | 56.08 | 46.46 | 51.22 | — | — | — | — | |
| Qwen2-VL-7BProcessing Mode=Offline, Params=7B2026.05 | 55.98 | — | — | — | — | — | — | — | — | — | 46.46 | — | 50.39 | — | 48.74 | — | |
| Dispider-7B#Frames=1 fps2026.04 | 54.6 | 57.7 | 49.5 | 62.1 | 44.9 | 61.4 | 51.6 | 48.5 | 55.4 | 4.3 | 36.1 | — | 45.35 | — | — | — | |
| Dispider-7B#Frames=1 fps2026.06 | 54.6 | — | — | — | — | — | — | — | — | — | 36.1 | — | 41.78 | 34.72 | — | 45.35 | |
| Dispider-7BModel Type=Online, Model Scale=7B2026.03 | 54.55 | 57.72 | 49.54 | 62.07 | 44.94 | 61.39 | 51.63 | 48.48 | 55.41 | 4.3 | 36.06 | 45.31 | — | — | — | — | |
| Dispider-7BProcessing Mode=Streaming, Params=7B2026.05 | 54.55 | — | — | — | — | — | — | — | — | — | 36.06 | — | 41.78 | — | 34.72 | — | |
| Dispider2026.05 | 54.5 | — | — | — | — | — | — | — | — | — | 36.1 | — | 41.8 | 34.7 | — | — | |
| ET-Instruct-3BProcessing Mode=Streaming, Params=3B2026.05 | 46.47 | — | — | — | — | — | — | — | — | — | 28.52 | — | 40.54 | — | 46.62 | — | |
| Flash-Vstream2026.05 | 29.9 | — | — | — | — | — | — | — | — | — | 25.4 | — | 33.2 | 44.2 | — | — | |
| Flash-VStream-7B#Frames=1 fps2026.04 | 28.4 | 24.2 | 29.4 | 28.5 | 33.7 | 25.7 | 28.8 | 39.1 | 37.2 | 5.9 | 27.4 | — | 27.9 | — | — | — | |
| Flash-VStream-7B#Frames=1 fps2026.06 | 28.4 | — | — | — | — | — | — | — | — | — | 27.4 | — | 33.61 | 45.09 | — | 27.9 | |
| Flash-VStream-7BModel Type=Online, Model Scale=7B2026.03 | 28.37 | 24.16 | 29.36 | 28.45 | 33.71 | 25.74 | 28.8 | 39.06 | 37.16 | 5.91 | 27.38 | 27.88 | — | — | — | — | |
| Flash-VStream-7BProcessing Mode=Streaming, Params=7B2026.05 | 28.37 | — | — | — | — | — | — | — | — | — | 27.38 | — | 33.61 | — | 45.09 | — | |
| VideoLLM-online-8B#Frames=2 fps2026.04 | 20.8 | 8.1 | 23.9 | 12.1 | 14 | 45.5 | 21.2 | 22.2 | 18.8 | 12.2 | 17.7 | — | 19.26 | — | — | — | |
| VideoLLM-online2026.05 | 20.8 | — | — | — | — | — | — | — | — | — | 17.7 | — | — | — | — | — | |
| VideoLLM-online-8B#Frames=2 fps2026.06 | 20.8 | — | — | — | — | — | — | — | — | — | 17.7 | — | — | — | — | 19.26 | |
| VideoLLM-online-8BModel Type=Online, Model Scale=8B2026.03 | 20.79 | 8.05 | 23.85 | 12.07 | 14.04 | 45.54 | 21.2 | 22.22 | 18.8 | 12.18 | 17.73 | 19.26 | — | — | — | — | |
| VideoLLM-online-8BProcessing Mode=Streaming, Params=8B2026.05 | 20.79 | — | — | — | — | — | — | — | — | — | 17.73 | — | — | — | — | — | |
| FluxMem-7BProcessing Mode=Streaming, Params=7B2026.05 | — | — | — | — | — | — | — | — | — | — | — | — | 53.4 | — | — | — |