Dialogue on URO-Bench-pro
69.1Understanding ScoreGemini-3.1 Pro
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemini-3.1 ProTrack=pro2026.04 | 69.1 | 84 | 99.2 | |
| Qwen3.5-Omni-PlusTrack=pro2026.04 | 66.3 | 86.3 | 99.8 | |
| Qwen3.5-Omni-FlashTrack=pro2026.04 | 64.1 | 83.8 | 98.7 |
Loading the SOTA2 catalog…