wan2.7-videoedit

Wan2.7-VideoEdit is a video editing model that allows users to make local or global modifications to existing videos through natural language instructions. Editing capabilities cover multiple dimensions: you can add, delete or replace video elements, switch scene environments (such as sunny to rainy days), change the overall visual style (such as realistic to watercolor, live-action to clay animation); character behavior, expressions and emotions can be adjusted, lines can be replaced and automatically match mouth shape and timbre; shooting methods can also be redefined, including parameters such as camera position, angle of view, scene range and focal length. The model also supports the replacement of video elements with reference images, and replicates dynamic processes such as actions, special effects, and camera movements in the video. The input modes are image, text and video, and the output is 720P or 1080P video with a duration of 2 to 10 seconds.

Context
0 tokens
Max output
0 tokens
Input
llm image video
Output
video

Pricing

Price: $0.15 / request

720P reference_to_video : $0.09999999999999999 /

720P reference_to_video : $0.09999999999999999 /

1080P reference_to_video : $0.15 /

1080P reference_to_video : $0.15 /

Endpoints

  • openai

More models

  • dreamina-seedance-2-0-mini-260615

    dreamina-seedance-2-0-mini 于 2026 年 6 月 15 日发布,是 Seedance 2.0 系列中最快、最实惠的版本,主打高性价比路线,价格约为标准版的 50%。尽管定位更轻量,其生成效果与 Seedance 2.0 Fast 较为接近,在部分场景下已能满足日常生产需求,并专门针对电商和广告场景进行了深度优化。模型支持原生音频、摄像机控制和图像或视频输入,输出分辨率限于 480P 和 720P,不支持 1080P。该模型保留了核心的参考生成能力,用户可通过融合提示词与最多 12 个多模态参考素材锁定人物一致性。

  • happyhorse-1.1-r2v
  • dreamina-seedance-2-0-260128

    Dreamina Seedance 2.0 是字节跳动 Seed 团队开发的音视频生成模型,并集成到 CapCut 中。该模型采用双分支扩散 Transformer 架构,在生成链路中实现视觉与听觉信息流的并行处理,通过注意力/桥接机制实现帧级精度的声画同步。模型支持文本、图像、音频和视频多模态输入,最高可同步解析 12 个参考文件,生成长达 15 秒的视频,具备原生音视频同步生成能力。模型配备了隐形水印和 C2PA 内容凭证等保护措施。

  • veo-3.1-generate-preview

    veo-3.1-generate-preview 是 Google Veo 3.1 的视频生成预览模型条目,适合将故事脚本、镜头描述或视觉创意转化为短视频原型。其价值在于帮助团队快速验证叙事节奏、场景氛围、动作表现和镜头构图,再进入后续剪辑或制作环节。高质量提示应包含主体特征、动作顺序、镜头景别、光线、风格、时长和禁用内容。应建立包含人物、运动、文字、复杂场景和不同画幅的样片集来评测,并在外部传播前完成版权、肖像、商标和安全审核。