X.PIN· @thexpin · X·· 3 小时前AI 评分65
AI 导读
据 The Decoder 报道，可灵（Kling）发布 Kling 4.0 视频模型，单条视频最长可生成 30 秒，提示词理解扩展到 8,000 tokens。新模型支持双声道立体声音频和超过十种语言的唇形同步，同时推出更轻量的 Flash 版本。
正文 · 原文
Kling AI released its Kling 4.0 video model, The Decoder reports. It generates single videos up to 30 seconds, extends prompt understanding to 8,000 tokens, and adds two-channel stereo audio and lip sync in more than ten languages. A lighter Flash variant is also out.
来源：X.PIN · x.com