List your company

General Language Capabilities on MMLU, GSM8K, GPQA, HumanEval, TruthfulQA, IFEval Aggregate

71.2Average ScoreGRPO
62.04864.42466.869.176May 26, 2025

Evaluation results

MethodLinks
GRPOBackbone=LLaMA-3.1-8B2025.0571.2
TI-DPOBackbone=LLaMA-3.1-8B2025.0571.1
TPOBackbone=LLaMA-3.1-8B2025.0570
CPOBackbone=LLaMA-3.1-8B2025.0568.9
KTOBackbone=LLaMA-3.1-8B2025.0568
DPOBackbone=LLaMA-3.1-8B2025.0566.8
TDPOBackbone=LLaMA-3.1-8B2025.0565.8
SFTBackbone=LLaMA-3.1-8B2025.0565.2
IPOBackbone=LLaMA-3.1-8B2025.0562.5
SIMPOBackbone=LLaMA-3.1-8B2025.0562.4