Image to Video
AI-generatedSummary
Generate a video from a specified image URL using selected Seedance models on the Kie.ai API.
Inputs
- Image URL (required) — URL of the source image to generate the video from.
- Prompt — Optional text prompt to guide the video generation.
- Model (required) — Select the Seedance model to use for image-to-video generation.
- Last Frame URL — Optional URL for the last video frame, applicable only for Seedance 2.0 models.
- Reference Image URLs — Optional up to 9 reference images for enhanced Seedance 2.0 multimodal context.
- Reference Video URLs — Optional up to 3 reference videos for Seedance 2.0 video style references.
- Reference Audio URLs — Optional up to 3 reference audio clips for Seedance 2.0 audio style references.
- Duration (required) — Select the desired video duration (4, 8, 12, or 15 seconds).
- Aspect Ratio (required) — Select the video's aspect ratio (e.g., 16:9, 9:16, 1:1, etc.).
- Resolution (required) — Choose the video resolution (480p or 720p), applicable for Seedance 2.0 models.
- Generate Audio (required) — Option to include audio generation in the video, for Seedance 2.0 models.
- Return Last Frame (required) — Option to return the last frame image separately, for Seedance 2.0 models.
- Web Search (required) — Option to enable web search to enhance generation, for Seedance 2.0 models.
- Seed — Numeric seed for random generation, set to 0 for random seed.
- Reply URL — Webhook URL to be called when the generation task completes.
- Reply Ref — Custom reference string passed back in the webhook callback.
- Negative Prompt — Text specifying elements to avoid in the generated video.
- Captcha Token — reCAPTCHA token if the API requires it for the request.
- Wait for Completion (required) — Whether to wait for the generation task to finish before returning (polls task status and waits up to 5 minutes).
Output shape
a single JSON object representing the creation task response or, if waited for completion, the final task result
If 'Wait for Completion' is enabled, the node polls for task completion and returns the completed task data. If not, it returns the initial creation response containing the task ID. The output includes URLs to generated video and optionally audio and frames, parsed for easy access.