Home>Strategy information>Software strategy

Zhiyuan releases the native multi-modal world model Emu3: realizing text, image and video understanding and generation only by predicting the next token

Author:Eve Cole Update Time:2024-12-03 16:48:01