List your company

Tool-use Mathematical Reasoning on Math

22.71Overall ScoreMENTOR
2.05567.417812.7818.1422Oct 21, 2025

Evaluation results

MethodLinks
MENTORModel=LLaMA-3.1-8B2025.1022.71
Strict RLModel=LLaMA-3.1-8B2025.1020.78
Outcome-based RLModel=LLaMA-3.1-8B2025.1020.42
SFTModel=LLaMA-3.1-8B2025.1018.06
VanillaModel=LLaMA-3.1-8B2025.1015.84
MENTORModel=LLaMA-3.2-3B2025.1011.27
Strict RLModel=LLaMA-3.2-3B2025.108.46
Outcome-based RLModel=LLaMA-3.2-3B2025.108.04
SFTModel=LLaMA-3.2-3B2025.105.62
VanillaModel=LLaMA-3.2-3B2025.102.85