Qwen2.5-Max

阿里巴巴云开发的大型语言模型系列

Qwen2.5-Max is a large-scale AI model using a mixture-of-experts (MoE) architecture. With extensive pre-training and fine-tuning, it delivers strong performance in benchmarks like Arena Hard, LiveBench, and GPQA-Diamond, competing with DeepSeek V3 and OpenAI.