List your company

Multi-subject Reasoning on MMLU-Pro

65.6Acc (Mean)Gemma-3-12B-Instruct
30.55239.65148.7557.849Dec 2, 2025

Evaluation results

MethodLinks
Gemma-3-12B-InstructSetting=Optimized2025.1265.60.01280.333102,069.6
Gemma-3-12B-InstructSetting=Random2025.1250.10.0080.263114,524.53
Qwen2.5-7B-InstructSetting=Optimized2025.1249.70.0180.174128,362.8
Qwen2.5-7B-InstructSetting=Random2025.1239.90.0150.114372,446.18
Llama-3.1-8B-InstructSetting=Optimized2025.1237.30.0110.155513,244.2
Llama-3.1-8B-InstructSetting=Random2025.1231.90.010.109822,103.05