美團發布1.6萬億參數智能體模型LongCat-2.5，不依賴英偉達GPU

meituan has released the weights for longcat-2.5, a 1.6-trillion-parameter mixture-of-experts model designed for ai agents that work through long, multi-step tasks autonomously. the model was trained on chinese hardware and performs close to top western models like claude opus 5 on several benchmarks, including coding tasks. longcat-2.5 activates only a subset of its specialized subnetworks for each token it processes, with roughly 48 billion active parameters per token. the model can handle up to one million tokens at once and generate responses up to 128,000 tokens long. it understands text, images, and video, and responds in text.