List your company

Question Answering on GPQA (avg.@8)

43.43Average Score (@8)Qwen2.5-7B-Instruct
18.41824.911531.40537.8985Aug 25, 2025

Evaluation results

MethodLinks
Qwen2.5-7B-InstructTraining Pipeline=PSFT → GRPO2025.0843.43
Qwen2.5-7B-InstructTraining Pipeline=SFT → GRPO2025.0839.39
Llama3.1-8B-InstructTraining Pipeline=PSFT → GRPO2025.0836.23
Qwen2.5-7B-InstructTraining Pipeline=PSFT2025.0833.21
Qwen2.5-7B-InstructTraining Pipeline=SFT2025.0832.89
Llama3.1-8B-InstructTraining Pipeline=SFT → GRPO2025.0831.06
Llama3.1-8B-InstructTraining Pipeline=PSFT2025.0826.89
Llama3.1-8B-InstructTraining Pipeline=SFT2025.0819.38