Calibration

ResearchClawBench → OSWorld 2.0

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

OSWorld 2.0 is estimated from ResearchClawBench with a offset logistic curve fitted on 5 models measured on both: y = 0.0719 + (1.2000 − 0.0719) / (1 + exp(−200.00·(x − 0.2211))), R² = 0.71, cross-validated error 5.1 pp. It is used for 2 estimates.

Estimated model	ResearchClawBench	OSWorld 2.0	Source
GLM-5.2	20.7%	13.6%	estimated ± 5.1 pp, low confidence
Grok 4.1	13.5%	7.2%	estimated ± 5.1 pp, low confidence