Calibration

OSWorld 2.0 → AA Agentic Index

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

AA Agentic Index is estimated from OSWorld 2.0 with a Hill curve fitted on 16 models measured on both: y = 0.0000 + (1.0483 − 0.0000)·x^0.40 / (0.80855^0.40 + x^0.40), R² = 0.74, cross-validated error 6.8 pp. It is used for 0 estimates.