Measured orbit coverage on T3Bench 300-prompt derived (frozen audit set)
359Metric A (Coverage Count/Score)
MedianGS-only R0
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.06 | 359 | 100 | 100 | |
2026.06 | 359 | 100 | 100 | |
2026.06 | 357 | 100 | 100 | |
2026.06 | 357 | 100 | 100 | |
2026.06 | 357 | 100 | 100 | |
2026.06 | 357 | 100 | 100 | |
2026.06 | 345 | 100 | 100 | |
2026.06 | 342.9 | 100 | 100 | |
2026.06 | 18.9 | 54 | 0 |