Seedance 2.5 Lite Reference to Video
Seedance 2.5 Lite Reference to Video is the special-price full-capability route that uses reference images, video, or audio to guide subjects, motion, visual style, and sound. It accepts up to 30 reference images, 10 reference videos, and 10 reference audios, supports durations up to 30 seconds, and outputs 720p only. This is a full-capability variant: media URLs are submitted to the asset review service before the task is submitted by default (real_person=true); set real_person=false only when the materials are confirmed face-free to pass media URLs through as-is, so plain public URLs always work in requests.
Asset and Scene Limits
Seedance 2.5 Lite reference-material mode is a special-price full-capability route with a built-in asset library, no facial distortion, a 100% pass rate for real-person content, and long-term stability. It bills at special-offer rates and only successful tasks are charged — see the pricing page for live rates.- Text prompt: For Chinese, keep it under 500 characters; for English, under 1,000 words.
- Image input: Supports jpeg, png, webp, bmp, tiff, gif, heic, and heif formats; each image under 30 MB; aspect ratio range (0.4, 2.5); width and height range (300px, 6000px).
- Image count: No more than 30 images.
- Video input: Each video 2–30 seconds; up to 10 reference videos; total video duration must not exceed 30 seconds.
- Audio input: Each clip 2–30 seconds; up to 10 reference audio clips; total audio duration must not exceed 30 seconds.
- Audio input restriction: Audio cannot be provided on its own; at least one reference image or reference video is required.
- Resolution: 720p only — 480p, 1080p, 2K and 4K are not supported on this route.
- Duration: 4–30 seconds per clip (default 4s).
Real-person review (real_person)
When a request carries reference materials, the booleanreal_person field controls whether materials are submitted to face review first:
real_person: true(default): materials are submitted to the asset review service first, and the task is created withassetId://{assetId}references; the task waits during review and fails if review rejects the materials.real_person: false: review is skipped and material URLs are passed through to task creation as-is.
false only when the materials are confirmed face-free to save the review round trip; keep the default true for materials containing real people.
Asset library references
Besides publicly accessible URLs, theimage_url / image_urls, video_urls, and audio_url / audio_urls fields also accept asset library references in the form assetId://{assetId}. Asset IDs are shared across the platform: any account may reference a known assetId, regardless of who submitted it — the same sharing semantics as a public URL. Only ACTIVE assets pass validation. Submit and track assets via asset upload and asset detail.Authorizations
Use your APIPod API key as a Bearer token in the Authorization header.
Body
Public APIPod model ID.
"seedance-2.5-lite-r2v"Generation or editing instructions, up to 30000 characters. It is recommended to keep the prompt to no more than 500 Chinese characters or 1,000 English words. Lengthy text will lead to scattered information, and the model may ignore details and only focus on key points, resulting in missing elements in the generated video.
30000Reference image URLs. Up to 30 images are supported; each image must be smaller than 30 MB.
30Reference video URLs. Up to 10 videos are supported; each video lasts 2-30 seconds and the total duration of all videos must not exceed 30 seconds.
10Reference audio URLs. Up to 10 audios are supported; each audio lasts 2-30 seconds and the total duration of all audios must not exceed 30 seconds. Please note that audio files cannot be uploaded alone; at least one reference video or image must be included.
10Video duration in seconds, from 4 to 30.
4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 4 <= x <= 30Generate video resolution.
720p Aspect ratio of the generated video,When the ratio is configured as adaptive, the model will automatically adapt the aspect ratio according to the generation scenario.
adaptive, 16:9, 4:3, 1:1, 3:4, 9:16, 21:9 Controls whether the generated video contains audio synchronized with the visuals.
Whether the reference materials contain real people (human faces). Defaults to true: materials are submitted to the asset review service and replaced with assetId:// references before task creation. Set to false only when the materials are confirmed face-free to skip review and pass material URLs through as-is.