Calibration
APEX-Agents-AA → ResearchClawBench
ResearchClawBench is estimated from APEX-Agents-AA with a linear curve fitted on 8 models measured on both: y = 0.1480·x + 0.1163, R² = 0.35, cross-validated error 2.6 pp. It is used for 1 estimate.
| Estimated model | APEX-Agents-AA | ResearchClawBench | Source |
|---|---|---|---|
| Claude Opus 4.6 (Adaptive) | 33.0% | 16.5% | estimated ± 2.6 pp, medium confidence |