| Benchmark | Model | Rank | Results |
|---|---|---|---|
| video-question-answering-on-streamingbench | MaLoW (Qwen3-VL-4B) | #2 | Real-Time Visual Understanding: 84.8 |
| video-question-answering-on-streamingbench | MaLoW (Qwen3.5-9B) | #6 | Real-Time Visual Understanding: 82.3 |
| video-question-answering-on-streamingbench | MaLoW (VideoChat3-4B) | #8 | Real-Time Visual Understanding: 81.2 |
| video-question-answering-on-streamingbench | MaLoW (MOSS-VL-Real-11B) | #9 | Real-Time Visual Understanding: 81.0 |
| robot-manipulation-on-simplerenv-widow-x | MaLoW (Qwen3.5-2B-MLP) | #21 | Average: 0.688 |