Calibration
ApprenticeBench → APEX-Agents-AA
APEX-Agents-AA is estimated from ApprenticeBench with a linear curve fitted on 5 models measured on both: y = 0.3878·x + 0.3182, R² = 0.46, cross-validated error 3.5 pp. It is used for 4 estimates.
| Estimated model | ApprenticeBench | APEX-Agents-AA | Source |
|---|---|---|---|
| Claude Fable 5 | 34.0% | 45.0% | estimated ± 3.5 pp, low confidence |
| Claude Opus 4.6 | 5.0% | 33.8% | estimated ± 3.5 pp, low confidence |
| Claude Sonnet 5 | 16.0% | 38.0% | estimated ± 3.5 pp, low confidence |
| Grok 4.6 | 13.0% | 36.9% | estimated ± 3.5 pp, low confidence |