AI 导读
Poolday 报告其视频智能体已累计完成 70 万次运行和 1 亿次视频编辑。该智能体接收视频需求,结合已有素材与品牌资产拼装出成片,需要新画面时可调用生成模型,例如用真实 UI 素材和 Figma 组件按品牌风格生成产品更新视频。输出按图层保持可编辑,可以更换 logo、修正字幕而保留视频其余部分。
正文
Ngl, thats really huge: Poolday reports 700,000 agent runs and 100 million video edits.
Its agent takes a video brief, works with your existing footage and brand assets, and assembles a finished cut. It can also use generative models when the video needs something new.
Say you're making a product update. You can use your real UI footage and actual Figma components, then have the agent put the video together in your brand's style.
And you still get to change your mind. The output stays editable, layer by layer. Swap the logo, fix a caption, keep the rest of the video.
来源:@kimmonismus · x.com