Calibration

ResearchClawBench → AA Agentic Index

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

AA Agentic Index is estimated from ResearchClawBench with a inverse Michaelis–Menten curve fitted on 10 models measured on both: y = 0.22222·(x − 0.0000) / (0.0000 + 0.3281 − x), R² = 0.68, cross-validated error 6.8 pp. It is used for 1 estimate.

Estimated model	ResearchClawBench	AA Agentic Index	Source
Grok 4.1	13.5%	15.5%	estimated ± 6.8 pp, medium confidence