About this model
SD 2.0 视频编辑是 ByteDance 推出的先进 AI 视频修改工具。它支持多镜头叙事,让你可以使用自然语言프롬프트无缝编辑和变换现有视频。无论是执行风格迁移、更换背景,还是修改场景中的特定元素,该模型都能在保持原始运动一致性和叙事结构的同时,提供导演级控制。
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.60 base price | muapiapp brings ByteDance's cutting-edge video editing to you early, providing premium video-to-video manipulation starting around $0.60 per edit depending on duration and quality. |
| Competitors | 조건부 변경 | 许多标准 API 缺少这一档位的原生视频转视频能力,因此 muapiapp 成为体验好莱坞级 AI 编辑的专属入口。 |
muapiapp brings ByteDance's cutting-edge video editing to you early, providing premium video-to-video manipulation starting around $0.60 per edit depending on duration and quality.
许多标准 API 缺少这一档位的原生视频转视频能力,因此 muapiapp 成为体验好莱坞级 AI 编辑的专属入口。
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| 프롬프트 | string | Text prompt describing the video edit. | Replace the running man with @image1. Preserve the exact running motion, speed, and camera shake. Ensure the armor glows dynamically with the environment lighting and reflects passing car lights. Maintain realistic foot contact with the ground and motion blur consistency. |
| 视频 URL | array | 最多上传 1 个비디오 URL。(最大大小:10MB,最大재생 시간:15s) | https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/seedance-v2.0-video-edit-in.avif |
| 이미지 URL | array | 最多上传 9 个이미지 URL。 | https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/seedance-v2.0-video-edit.jpg |
| 参考오디오 URL | array | Up to 3 reference audio clip URLs (MP3/WAV, total max 15s). Each Nth audio corresponds to @audioN in the prompt. | undefined |
| 화면 비율 | Enum (4 options) | - | 16:9 |
| 품질 | Enum (2 options) | - | basic |
| 재생 시간 (초)(秒) | int | Output video duration in seconds (4–15). | 5 |
Text prompt describing the video edit.
Replace the running man with @image1. Preserve the exact running motion, speed, and camera shake. Ensure the armor glows dynamically with the environment lighting and reflects passing car lights. Maintain realistic foot contact with the ground and motion blur consistency.最多上传 1 个비디오 URL。(最大大小:10MB,最大재생 시간:15s)
https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/seedance-v2.0-video-edit-in.avif最多上传 9 个이미지 URL。
https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/seedance-v2.0-video-edit.jpgUp to 3 reference audio clip URLs (MP3/WAV, total max 15s). Each Nth audio corresponds to @audioN in the prompt.
undefined-
16:9-
basicOutput video duration in seconds (4–15).
5Developer documentation
上传源视频:通过 video_urls 数组提供目标视频的有效 URL(最多 1 个视频)。
(可选)添加参考图像:如果希望模型在编辑过程中模仿特定的美学、角色或物体风格,请通过 images_list 添加最多 9 个图像 URL。
编写编辑프롬프트:清楚描述视频中应该发生什么。如果上传了参考图像,请使用 @image1、@image2 引导模型(例如:'将环境变成 @image1 中的城市')。
选择时长:选择출력 결과视频长度(4–15s)。
选择质量设置:选择 basic(출력 결과 $0.21/sec,입력 파라미터视频每秒另收 $0.063/sec)以获得更快、更经济的渲染,或选择 high(출력 결과 $0.30/sec,입력 파라미터视频每秒另收 $0.09/sec)以获得最高视觉保真度。입력 파라미터源视频会根据其时长收取 30% 附加费。
提交请求:提交后,模型会출력 결과与프롬프트匹配的无缝编辑视频。
Frequently asked
目前,SD 2.0 视频编辑模型严格专注于视觉变换。音频处理通常需要单独完成。
为获得最佳效果并符合 API 处理限制,建议提供时长少于 15 秒的源视频。
'basic' 使用更快的渲染流程,适合快速原型制作;'high' 使用标准的深度渲染流程,适合生产环境和电影级质量。