Agent tool-use on ClawEval 114 T-family tasks
37.7p@1
SFT ISETrace
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.06 | 37.7 | 53.3 | 95.9 | |
2026.06 | 30.7 | 44.6 | 94.7 | |
2026.06 | 25.4 | 44.2 | 96.5 | |
2026.06 | 19.3 | 36.7 | 92.5 |
SFT ISETrace
| Method | Links | |||
|---|---|---|---|---|
2026.06 | 37.7 | 53.3 | 95.9 | |
2026.06 | 30.7 | 44.6 | 94.7 | |
2026.06 | 25.4 | 44.2 | 96.5 | |
2026.06 | 19.3 | 36.7 | 92.5 |