| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| WebArena (test) | APEX | Shopping Success Rate42.9 | 18 | 2mo ago | |
| HLE | SlimSearcher (SFT+RL) | Accuracy37.6 | 15 | 1mo ago | |
| Browsecomp | Accuracy51.5 | 15 | 1mo ago | ||
| GAIA | SlimSearcher (SFT+RL) | Accuracy70.9 | 14 | 1mo ago | |
| XBench DeepSearch | SlimSearcher (SFT+RL) | Accuracy79 | 14 | 1mo ago | |
| GoBrowse | CATTS (∆, best τ) | Success Rate90.4 | 4 | 4mo ago | |
| WebArena Lite | CATTS (H, best τ) | Success Rate0.479 | 4 | 4mo ago |