Calibration
Toolathlon → ApprenticeBench
ApprenticeBench is estimated from Toolathlon with a Hill curve fitted on 5 models measured on both: y = 0.0000 + (1.2000 − 0.0000)·x^6.00 / (0.74841^6.00 + x^6.00), R² = 0.52, cross-validated error 5.8 pp. It is used for 0 estimates.
Fitted on
| Model | Toolathlon | ApprenticeBench |
|---|---|---|
| GPT-5.4 | 54.6% | 11.0% |
| GPT-5.5 | 55.6% | 20.0% |
| GPT-5.6 Luna | 53.4% | 7.0% |
| GPT-5.6 Sol | 58.0% | 26.0% |
| GPT-5.6 Terra | 53.1% | 16.0% |