Calibration

MCP Atlas → Toolathlon

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

Toolathlon is estimated from MCP Atlas with a Hill curve fitted on 17 models measured on both: y = 0.3617 + (1.2000 − 0.3617)·x^6.00 / (0.95188^6.00 + x^6.00), R² = 0.72, cross-validated error 7.5 pp. It is used for 2 estimates.

Estimated model	MCP Atlas	Toolathlon	Source
LLaDA2.2-flash	46.2%	37.2%	estimated ± 7.5 pp, medium confidence
Solar Open 2	58.2%	40.3%	estimated ± 7.5 pp, medium confidence