benchgap
Calibration

Toolathlon → WideResearch

WideResearch is estimated from Toolathlon with a linear curve fitted on 7 models measured on both: y = 0.6351·x + 0.4879, R² = 0.65, cross-validated error 5.6 pp. It is used for 5 estimates.

Estimated modelToolathlonWideResearchSource
Gemini 3.5 Flash56.5%84.7%estimated ± 5.6 pp, low confidence
GLM-5.248.2%79.4%estimated ± 5.6 pp, low confidence
GPT-5.4 mini42.9%76.0%estimated ± 5.6 pp, low confidence
GPT-5.4 nano35.5%71.3%estimated ± 5.6 pp, low confidence
MiniMax M2.746.3%78.2%estimated ± 5.6 pp, low confidence