For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Animate a still image with realistic motion, depth, and audio. Provide a reference image plus a prompt describing the desired motion. The output preserves the visual identity of the source image.
Responds immediately with the job `id` and `created_at` timestamp. Poll `GET /v2/image-to-video/{id}` until the status is `completed`, then download from `result.video_url`.
Authentication
AuthorizationBearer
API key authentication
Request
This endpoint expects an object.
image_uristringRequired
Image to be used as the first frame of the video. See Input Formats for supported formats and size limits.
promptstringRequired<=5000 characters
Text description of how the image should be animated, can relate to the details in the image
Video duration in seconds. See Supported Models for available durations per model.
resolutionstringRequired
Output video resolution. See Supported Models for available resolutions per model.
fpsintegerOptionalDefaults to 24
Frame rate in frames per second. See Supported Models for available FPS per model and resolution.
last_frame_uristringOptional
Image to be used as the last frame of the video. When provided, the video will interpolate between the first frame (image_uri) and this last frame. See Input Formats for supported formats and size limits.
Note: Only supported by ltx-2-3 models (ltx-2-3-fast, ltx-2-3-pro).
camera_motionenumOptional
Apply camera motion effects to the generated video.
generate_audiobooleanOptionalDefaults to trueBeta
Generate audio for the video. When true, the generated video will include AI-generated audio matching the scene. When false, only silent video is generated.
Response
Job submitted successfully
idstring
Unique job identifier. Use this to poll for status.
Animate a still image with realistic motion, depth, and audio. Provide a reference image plus a prompt describing the desired motion. The output preserves the visual identity of the source image.
Responds immediately with the job id and created_at timestamp. Poll GET /v2/image-to-video/{id} until the status is completed, then download from result.video_url.
