benchgap
Calibration

AA-AnalystAgent → Harvey LAB

Harvey LAB is estimated from AA-AnalystAgent with a Michaelis–Menten curve fitted on 10 models measured on both: y = 0.9512·x / (0.00935 + x), R² = 0.90, cross-validated error 1.8 pp. It is used for 2 estimates.

Estimated modelAA-AnalystAgentHarvey LABSource
GPT-6.1 Sol (max)50.0%93.4%estimated ± 1.8 pp, high confidence
Gemini 3.7 Flash (high)60.0%93.7%estimated ± 1.8 pp, medium confidence