创建 React 动态图形、带字幕的解说视频和可复用的视频合成,并导出为 MP4。
将简报、脚本、视觉参考和提供的素材转换为基于 React 的视频合成,并渲染为 MP4。使用 Remotion 在 video-base 沙盒中处理场景计时、动态效果、字幕和音频驱动的布局。
按需使用平台旁白和音乐生成功能时,会产生平台费用。视觉内容通过代码和提供的素材制作。字幕需要提供时间信息或经验证的音频分析。额外的 3D、Lottie、特效包、外部服务和交互式 Studio 访问取决于运行时支持,不包含在基础承诺中。
从英语翻译, 仅供参考.
Generate a video introducing the most powerful AI models in the world today (2 or 3 of them), around 10–20 seconds long.
I'll start by loading the Remotion skill to get the platform-specific workflow.
Now I'll build the Remotion project. First, the fonts and theme:
Your video is ready — a 19.8-second motion-graphics intro to the three models at the top of the frontier right now.
What it covers
Each card has its own animated signature graphic (terracotta bloom, mint hex lattice, violet orb), a count-up score, and a source footnote. The soundtrack is a custom cinematic bed mixed under a narration track.
Verified output: 1920×1080, 30 fps, 593 frames, 19.77 s, H.264 + AAC stereo; 593 frames decoded, audio levels checked (voice peaks −2.8 dB, music bed ~10 dB below narration), and first/mid/final frames inspected for type sharpness, overflow, and correct values.
One caveat worth stating plainly: the scores come from an Artificial Analysis Intelligence Index snapshot for October 2026, and rankings shift week to week — the video's closing point, that no single model wins everything, is the accurate takeaway.
本次花费400 积分≈ $0.400文字转语音 ×8、音乐生成 ×1、网页搜索 ×1DeepSeek V4.1 Flash 1017 · 3,340,831 tokens
运行智能体会消耗你的积分。可随时停止——你只需为它实际产出的内容付费。