Video Generation (Coming soon)
Create text-to-video, image-to-video, and video-to-video tasks.
Video generation (Coming soon)
Use POST /v1/videos/generations to create a video task. The API returns a task id; use Task status to get the final video URL.
This endpoint is documented for preview and is not open yet.
Endpoint
POST /v1/videos/generationsRequest body
| Field | Type | Required | Description |
|---|---|---|---|
provider | string | Yes | Video provider, for example kie or fal. |
model | string | Yes | Video model, for example sora-2-pro-text-to-video. |
scene | string | Yes | One of text-to-video, image-to-video, or video-to-video. |
prompt | string | Required for text-to-video | Text instruction for the video. |
image_input | string or string[] | Required for image-to-video | One or more image URLs used as references. |
video_input | string or string[] | Required for video-to-video | Source video URL used as the input. |
options | object | No | Extra model options, such as duration or aspect_ratio when the selected model supports them. |
Scenes and credits
| Scene | Required input | Credits |
|---|---|---|
text-to-video | prompt | 6 |
image-to-video | image_input | 8 |
video-to-video | video_input | 10 |
Example: text to video
curl https://sseedance.com/v1/videos/generations \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"provider": "kie",
"model": "sora-2-pro-text-to-video",
"scene": "text-to-video",
"prompt": "A cinematic shot of a glass city at sunrise"
}'Example: image to video
curl https://sseedance.com/v1/videos/generations \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"provider": "kie",
"model": "sora-2-pro-image-to-video",
"scene": "image-to-video",
"image_input": ["https://cdn.example.com/reference.png"],
"options": {
"duration": "10"
}
}'Example: video to video
curl https://sseedance.com/v1/videos/generations \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"provider": "fal",
"model": "fal-ai/kling-video/o1/video-to-video/edit",
"scene": "video-to-video",
"video_input": ["https://cdn.example.com/source.mp4"]
}'Response
{
"id": "task_id",
"object": "video.generation.task",
"status": "pending",
"data": {
"id": "task_id",
"mediaType": "video",
"status": "pending",
"videoUrls": []
}
}Use GET /v1/tasks/{id} to check the final videoUrls.