Qwen2.5 Technical Report

Open paper Other version

Benchmark	Model	Rank	Results
code-generation-on-humaneval	Qwen2.5-Plus	–	Pass@1: 87.8
mathematical-reasoning-on-aime24	Qwen2.5-72B-Instruct	#5	Acc: 23.3
question-answering-on-gpqa	Qwen2.5-72B-Instruct	#4	Accuracy: 49