Proactive Task Management on IPIBench
98Cancel ScoreHuman Level
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Human Level2026.05 | 98 | 92 | 98 | 96 | |
| GPT-5.4Model Type=Proprietary, Inference Mode=Offline2026.05 | 38.43 | 24.5 | 7.57 | 23.5 | |
| Gemini 2.5 ProModel Type=Proprietary, Inference Mode=Offline2026.05 | 24.54 | 20.81 | 16.78 | 20.71 | |
| Gemini 3 ProModel Type=Proprietary, Inference Mode=Offline2026.05 | 19.44 | 21.84 | 19.39 | 20.22 | |
| GLM-4.6VModel Type=Open-source, Inference Mode=Offline2026.05 | 15.28 | 24.16 | 9.69 | 16.38 | |
| Qwen3.5-PlusModel Type=Open-source, Inference Mode=Offline2026.05 | 12.96 | 31.54 | 10.87 | 18.46 | |
| GPT-4oModel Type=Proprietary, Inference Mode=Offline2026.05 | 11.57 | 31.88 | 14.18 | 19.21 | |
| VideoLLM-online-8BModel Type=Open-source, Inference Mode=Online2026.05 | 7.87 | 1.68 | 6.62 | 5.39 | |
| InternVL3-8BModel Type=Open-source, Inference Mode=Offline2026.05 | 6.94 | 6.04 | 3.55 | 5.51 | |
| Flash-VStream-7BModel Type=Open-source, Inference Mode=Online2026.05 | 2.31 | 0 | 0 | 0.77 | |
| LLaVA-OneVision-7BModel Type=Open-source, Inference Mode=Offline2026.05 | 1.39 | 1.01 | 0.47 | 0.96 | |
| Qwen3-VL-8BModel Type=Open-source, Inference Mode=Offline2026.05 | 1.39 | 17.45 | 9.22 | 9.35 |