List your company

Reasoning on MMLU-Pro (ER, TP %)

3ERB-PAC
2.645.077.59.93Jan 30, 2026

Evaluation results

MethodLinks
B-PACRisk tolerance (epsilon)=0.08, Confidence level (alpha)=0.12026.01377.24
CoDRisk tolerance (epsilon)=0.08, Confidence level (alpha)=0.12026.011073.63
NoThinkingRisk tolerance (epsilon)=0.08, Confidence level (alpha)=0.12026.011276.52