| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multimodal Deep Search | HLE-VL | Accuracy (HLE-VL)19 | 18 | |
| Multimodal long-horizon complex reasoning | HLE-VL | Score25.1 | 13 | |
| Multimodal Search | HLE-VL | Score36 | 8 | |
| Vision Agent Capacity | HLE-VL | Score39.2 | 6 | |
| Visual Search | HLE-VL | Pass@136 | 4 |