Veo 3.1 参考图转视频: Image-to-Video
Veo 3.1 R2V 可让创作者使用最多三张参考图像生成动态视频。该模型在整个视频中保持角色、物体和风格的视觉一致性，生成电影级的 8 秒片段。它非常适合将概念艺术、分镜或角色设计转换为短动画序列，同时保留原始美学风格。
Veo 3.1 R2V 可让创作者使用最多三张参考图像生成动态视频。该模型在整个视频中保持角色、物体和风格的视觉一致性，生成电影级的 8 秒片段。它非常适合将概念艺术、分镜或角色设计转换为短动画序列，同时保留原始美学风格。
About this model
Veo 3.1 R2V 是一款创新模型，可将最多三张参考图像转换为动态、电影级视频片段。借助先进的Imagem para Vídeo技术，它能够确保角色、物体和风格在整个 8 秒动画序列中保持视觉一致性。无论你是在开发概念艺术、分镜还是细致的角色设计，该模型都能将静态图像转换为流畅的叙事体验，同时保留原始Entrada的艺术完整性。
该模型基于强大的底层 AI 技术，将深度学习算法与视频合成能力相结合，在 720p 或 1080p 下Saída高分辨率视频。它能够智能生成音频，进一步提升观众体验，因此非常适合希望制作引人入胜的视觉故事和宣传内容的创作者。这种精准执行与创作灵活性的结合，使该模型成为现代多媒体制作中具有竞争力的解决方案。
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.6 | Offers exceptional value by being 20-50% more affordable than competitors while delivering comparable or superior quality. |
| Fal.ai | $0.75 | Priced at $0.75 per generation, which is 20-50% higher than muapiapp, yet provides similar video quality. |
| Replicate | $0.75 | Matches Fal.ai's pricing, making muapiapp a more cost-effective option at 20-50% lower cost with similar or better output. |
Offers exceptional value by being 20-50% more affordable than competitors while delivering comparable or superior quality.
Priced at $0.75 per generation, which is 20-50% higher than muapiapp, yet provides similar video quality.
Matches Fal.ai's pricing, making muapiapp a more cost-effective option at 20-50% lower cost with similar or better output.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | 用于生成Vídeo的Prompt | A small robotic fox exploring a sun-drenched enchanted forest. The fox hops across a sparkling stream, pauses on mossy rocks, and looks curiously at glowing fireflies. Cinematic camera pans follow the fox from behind, then orbit slightly to reveal sunbeams filtering through the canopy. Warm dappled lighting with volumetric light rays and soft particle effects. Gentle ambient forest sounds and faint magical chimes. Dialogue: ‘Everything shines differently under the forest light…’ |
| Imagem URL | array | 上传或提供Imagem URL，用于Imagem转Vídeo生成。 | https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/veo3.1-reference-to-video-1.jpg |
| Resolução | Enum (3 options) | 生成Vídeo的Resolução。 | 720p |
| Duração (segundos) | Enum (1 options) | 生成Vídeo的Duração（秒）。 | 8 |
| 生成Áudio | boolean | 是否生成Áudio。 | true |
用于生成Vídeo的Prompt
A small robotic fox exploring a sun-drenched enchanted forest. The fox hops across a sparkling stream, pauses on mossy rocks, and looks curiously at glowing fireflies. Cinematic camera pans follow the fox from behind, then orbit slightly to reveal sunbeams filtering through the canopy. Warm dappled lighting with volumetric light rays and soft particle effects. Gentle ambient forest sounds and faint magical chimes. Dialogue: ‘Everything shines differently under the forest light…’上传或提供Imagem URL，用于Imagem转Vídeo生成。
https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/veo3.1-reference-to-video-1.jpg生成Vídeo的Resolução。
720p生成Vídeo的Duração（秒）。
8是否生成Áudio。
trueDeveloper documentation
Como Usar Veo 3.1 R2V
准备Entrada
提交请求
veo3.1-reference-to-video。prompt 和 images_list。generate_audio 设置为 true 或 false，决定是否生成音频。查看并解读结果
Frequently asked
模型需要一段详细的文本Prompt，以及最多三张以 URL 或上传方式提供的参考图像。此外，还可以设置分辨率、时长和是否生成音频等参数。
模型生成电影级的 8 秒视频片段，并支持 720p 和 1080p 视频分辨率。
可以。提交Entrada时，将 `generate_audio` 参数设置为 `true`，即可同时生成音频。
Veo 3.1 R2V 专注于保持角色和物体的视觉一致性，确保风格与美学特征忠实于原始参考素材。音频生成功能也为最终视频制作增加了更多层次，使其成为适合创意项目的综合工具。