跳到正文
Jason· 来源· @jack_gor · X·· 7 天前精选AI 评分82
推荐理由

原文披露了音频支持与场景时长翻倍等关键能力变化,读者可据此评估其技术演进与可用性。

正文 · 原文

DeepMind releases Genie 4 research preview: real-time interactive world model with audio, maintaining single scenes for ~2 mins (2x Genie 3). Currently limited to trusted testers; coming to Google AI Ultra subscribers in weeks. Known issues: detail drift and unreadable text.
https://techcrunch.com/2026/10/27/googles-genie-4-world-model-adds-audio-more-realistic-worlds-and-real-time-steering/

来源:Jason · x.com