新浪科技讯 3月27日上午消息，美团发布并全面开源原生多模态大模型LongCat-Next及其核心组件离散原生分辨率视觉分词器（dNaViT）。

　　该模型打破了当前大模型以“语言为中心”的传统拼凑式架构，将图像、语音与文本统一映射为同源的离散Token。通过纯粹的“下一个Token预测”（Next Token Prediction，NTP）范式，LongCat-Next让视觉与语音成为AI的“原生母语”。

　　据介绍，LongCat-Next实现了三项关键技术突破：一是离散原生自回归架构（DiNA）彻底打破模态隔阂；二是离散原生分辨率视觉分词器（dNaViT）构造视觉世界的“词典”，三是语义对齐完备编码器破解“离散化必然损失信息”的行业难题。

海量资讯、精准解读，尽在新浪财经APP

责任编辑：江钰涵

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.