阶跃星辰联合吉利汽车集团开源两款阶跃Step系列多模态大模型

Core Insights - The collaboration between Jiyue Star and Geely Automobile Group has led to the open-source release of two multimodal large models: Step-Video-T2V for video generation and Step-Audio for voice interaction [1] Group 1: Model Performance - The Step-Video-T2V model is reported to have a leading parameter count and performance in the global open-source video generation field [1] - Step-Audio is noted as the industry's first product-level open-source voice interaction model [1]