Modellix

google/veo-3-i2v

文档
Schema
Google's image-to-video model for animating a starting frame, with an optional end frame and native sound. It makes a 4, 6, or 8 second clip at 720p or 1080p.模型说明收起说明
$0.4600/sec
image-to-video

输入

Video description text guiding the animation
Starting frame. Accepts an image URL (e.g. `https://example.com/frame.jpg`). Required for I2V.
提示:可拖拽文件、从剪贴板粘贴(Ctrl/Cmd+V),或提供 URL。
Text describing what to avoid in the generated video.
Video aspect ratio
Video duration in seconds (string type)
End frame. Accepts an image URL. Optional, guides the video to end at this frame.
提示:可拖拽文件、从剪贴板粘贴(Ctrl/Cmd+V),或提供 URL。
Person generation policy
Video resolution

结果

暂无结果

运行模型后,结果将在这里显示。

Next:

README

属性

系列

参数

参数名 描述 类型 必填 枚举值
prompt Video description text guiding the animation string 是 -
image Starting frame. Accepts an image URL (e.g. https://example.com/frame.jpg). Required for I2V. string 是 -
negativePrompt Text describing what to avoid in the generated video. string 否 -
aspectRatio Video aspect ratio string 否 16:9, 9:16
duration Video duration in seconds (string type) string 否 4, 6, 8
lastFrame End frame. Accepts an image URL. Optional, guides the video to end at this frame. string 否 -
personGeneration Person generation policy string 否 allow_adult
resolution Video resolution string 否 720p, 1080p

价格

输入图片($/img): 0.0000
输入视频($/sec): 0.0000

单位: $/sec

价格
$0.4600/sec

相关模型

  • google/veo-3.1-lite-i2v: Google’s image-to-video model for animating a still or bridging a start and end frame, with native sound. It makes a 4, 6, or 8 second clip at 720p or 1080p, in widescreen or vertical frames.
  • google/veo-3.1-lite-t2v: Google’s efficient text-to-video model for high-volume social clips and ads with native sound. It makes a 4, 6, or 8 second clip at 720p or 1080p, in widescreen or vertical frames.
  • google/veo-3.1-fast-i2v: Google’s fast image-to-video model for animating a still, bridging two frames, or matching up to three reference images, with native sound. It makes a 4, 6, or 8 second clip in widescreen or vertical frames, and can render 1080p or 4K.
  • google/veo-3.1-i2v: Google’s cinematic image-to-video model for animating a still, bridging two frames, or following up to three reference images, with native sound. It makes a 4, 6, or 8 second clip in widescreen or vertical frames, and can render 1080p or 4K.
  • google/veo-3.1-fast-t2v: Google’s fast text-to-video model for drafts, social clips, and high-volume ads with native sound. It makes a 4, 6, or 8 second clip in widescreen or vertical frames, and can render 1080p or 4K.
  • google/veo-3.1-t2v: Google’s cinematic text-to-video model for short films, ads, and scenes with native sound. It makes a 4, 6, or 8 second clip in widescreen or vertical frames, and can render 1080p or 4K.

相关资源