Streaming Question Answering on OVO-Bench Overall
92.81Average ScoreHuman
Evaluation results
| Method | Links | |
|---|---|---|
| Human2026.06 | 92.81 | |
| Gemini 1.5 ProResponse Strategy=Fixed decoding2026.06 | 63 | |
| GPT-4oResponse Strategy=Fixed decoding2026.06 | 59.54 | |
| Qwen2.5-VLResponse Strategy=Fixed decoding2026.06 | 54.2 | |
| InternVideo2.5Response Strategy=Fixed decoding2026.06 | 53.21 | |
| LLaVA-VideoResponse Strategy=Fixed decoding2026.06 | 53.1 | |
| MiniCPM-V2.6Response Strategy=Fixed decoding2026.06 | 51.47 | |
| LyraVResponse Strategy=Proactive Response2026.06 | 50.97 | |
| LiveStarResponse Strategy=Proactive Response2026.06 | 50.34 | |
| DispiderResponse Strategy=Proactive Response2026.06 | 41.78 | |
| MMDuetResponse Strategy=Proactive Response2026.06 | 39.85 | |
| VideoLLM-onlineResponse Strategy=Proactive Response2026.06 | 12.8 |