Latest Search
Quote
| Back Zoom + Zoom - | ||||
|
MEITUAN-W Open-Sources Native Multimodal Large Model LongCat-Next
Recommend 12 Positive 21 Negative 5 |
||||
|
||||
|
MEITUAN-W (03690.HK) released and fully open-sourced the native multimodal large model LongCat-Next. According to the introduction, this model breaks the traditional language-centric patchwork architecture of current large models by uniformly mapping images, voice and text to discrete tokens of the same origin. Through a purely "next token prediction" paradigm, the new model allows visual and speech to become the "native languages" of AI, representing a step forward for the Meituan LongCat team on the path to AI in the physical world, based on introduction. AASTOCKS Financial News Website: www.aastocks.com |
||||
Copied to Clipboard
 |
||||
