| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| SkillCraft Hard | Minimax-M2.1 | Success Rate20 | 16 | 1mo ago | |
| SkillCraft Overall | GPT-5.2 | Avg Tokens (M)0.26 | 16 | 1mo ago | |
| MCP Tool-Use Tasks 208 tasks from 52 tools (evaluation) | TRUSTDESC | Task Success Rate (TSR)89.9 | 11 | 9d ago | |
| Tau-Bench Airline v2 (test) | Pass Rate70 | 11 | 13d ago |