Seedance 2.0 video generation API
seedance2.0 uses the shared async task API and supports text-to-video, paired first and last frames, and reference-material generation.
Generation modes
| Mode | Rule |
|---|---|
| Text to video | references is empty or omitted |
| First and last frame | One first_frame and one last_frame image |
| Reference to video | One or more role=reference materials |
Reference limits
| Material | Limit |
|---|---|
| Reference images | 9 |
| Reference videos | 3 |
| Reference audio | 3 |
Requests above the limit return HTTP 400. Reference images, video, or audio cannot be mixed with first/last-frame mode.
curl --request POST "https://ling-light.com/v1/videos" \
--header "Authorization: Bearer $VIDEO_API_KEY" \
--header "Content-Type: application/json" \
--header "Idempotency-Key: $(uuidgen)" \
--data '{
"model": "seedance2.0",
"prompt": "An orange cat watching rain from a windowsill, slow camera push-in",
"seconds": 5,
"resolution": "720p",
"aspect_ratio": "16:9"
}'{
"model": "seedance2.0",
"prompt": "The person moves naturally from @图片1 to @图片2 while the camera stays stable",
"seconds": 5,
"resolution": "720p",
"aspect_ratio": "16:9",
"references": [
{
"type": "image",
"role": "first_frame",
"source": "https://cdn.example.com/first-frame.png"
},
{
"type": "image",
"role": "last_frame",
"source": "https://cdn.example.com/last-frame.png"
}
]
}First and last frame guide
Learn order, prompt references, and pairing rules.
Seedance 2.5
Compare the higher reference limits.