| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MCP-Universe | PREPING | Financial Success Score70.8 | 41 | 1mo ago | |
| MiniHack | ALMA | Success Rate20 | 12 | 4mo ago | |
| Baba Is AI | Dynamic Cheatsheet | Success Rate38 | 12 | 4mo ago | |
| Textworld | ALMA | Success Rate75 | 12 | 4mo ago | |
| ALFWorld | ALMA | Success Rate87.1 | 12 | 4mo ago | |
| ScienceWorld full (val) | TREK (DeepSeek-V4) | Success Rate26.7 | 4 | 18d ago | |
| ALFWorld (val) | TREK (DeepSeek-V4) | Success Rate82.8 | 4 | 18d ago |