LTX Video 2 Standard imagen a vídeo: Image-to-Video
LTX-2-19B Image-to-Video 可将单张图像制作成连贯的电影感片段，具备出色的时间稳定性。它保留构图和光照，同时加入受控的镜头运动、逼真的视差和细腻的环境动态，非常适合扎实的场景、近未来概念和关键故事节拍。
LTX-2-19B Image-to-Video 可将单张图像制作成连贯的电影感片段，具备出色的时间稳定性。它保留构图和光照，同时加入受控的镜头运动、逼真的视差和细腻的环境动态，非常适合扎实的场景、近未来概念和关键故事节拍。
About this model
LTX-2-19B Image-to-Video 是一款前沿 AI 模型，能够将单张图像转换为动态的电影感片段。借助先进的深度学习技术，该模型在保留原始构图和光照的同时，确保出色的时间稳定性。它支持受控的镜头运动、逼真的视差效果和细腻的环境动态，可以创作沉浸式视觉叙事，提升故事表达体验。
LTX-2-19B Image-to-Video 专为扎实的场景、近未来概念和精准的故事节拍而设计，擅长生成自然且技术感十足的高质量动画。其底层算法注重保持电影感节奏和城市氛围，是希望将静态图像变为连贯且富有视觉吸引力的视频salida的艺术家、创意专业人士和内容创作者的理想方案。
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.6 per generation | muapiapp offers this service at $0.6 per generation, which is 20-50% more affordable than competing providers while delivering comparable or superior quality. |
| Fal.ai | $0.9 per generation | Fal.ai charges approximately $0.9 per generation, making muapiapp 20-50% more cost-effective with equally robust performance. |
| Replicate | $0.9 per generation | Replicate also offers similar capabilities at around $0.9 per generation. muapiapp remains 20-50% cheaper while ensuring top-notch output quality. |
muapiapp offers this service at $0.6 per generation, which is 20-50% more affordable than competing providers while delivering comparable or superior quality.
Fal.ai charges approximately $0.9 per generation, making muapiapp 20-50% more cost-effective with equally robust performance.
Replicate also offers similar capabilities at around $0.9 per generation. muapiapp remains 20-50% cheaper while ensuring top-notch output quality.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | 描述vídeo内容的文本prompt。 | Animate the scene so the camera slowly pushes toward the billboard, the text characters on the woman’s face subtly scrolling and re-forming, rain falling continuously, reflections on the wet road shifting as car headlights flicker, pedestrians making small natural movements while the city lights pulse softly; maintain realistic motion, urban mood, and cinematic pacing. |
| Imagen URL | string | entradaimagen的 URL。 | https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/ltx-2-19b-i2v.jpg |
| Duración (segundos) | int | 生成vídeo的duración（秒）。 | 5 |
| Resolución | Enum (3 options) | 生成vídeo的resolución。 | 720p |
描述vídeo内容的文本prompt。
Animate the scene so the camera slowly pushes toward the billboard, the text characters on the woman’s face subtly scrolling and re-forming, rain falling continuously, reflections on the wet road shifting as car headlights flicker, pedestrians making small natural movements while the city lights pulse softly; maintain realistic motion, urban mood, and cinematic pacing.entradaimagen的 URL。
https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/ltx-2-19b-i2v.jpg生成vídeo的duración（秒）。
5生成vídeo的resolución。
720pDeveloper documentation
准备entrada
提交请求
prompt：视频场景和运动细节的描述。image_url：entrada图像的 URL。resolution：所需的视频质量。duration：视频时长（以秒为单位）。接收并查看视频
video URL 获取视频。集成与分享
Frequently asked
模型接受指向 JPEG 和 PNG 等标准图像格式的图像 URL。为了获得最佳结果，请确保entrada图像质量良好。
你可以在entrada JSON 中指定时长（以秒为单位）和分辨率（480p、720p、1080p）。时长必须介于 5 到 20 秒之间，默认值为 5 秒。
可以。你可以通过 `prompt` 参数详细描述所需的镜头运动和环境动态，从而控制场景在视频中的演变方式。
每次生成的费用为 $0.6，与其他市场提供商相比，这是一种高性价比的选择。