Bytedance · Video

Seedance 2.0 Fast Reference-to-Video

Realistic videoProduct shots

Seedance 2.0 Fast Reference to Video creates cinematic videos with native audio from your own reference images, videos, and audio, keeping characters, products, motion, and sound consistent while a prompt directs the new shot.

from€0.40per render

Examples

Seedance 2.0 Fast Reference-to-Video: Use the woman and tin soldier from <Image 1>. Preserve her exact face, glasses, clothing, hand position, toy d
Prompt

Use the woman and tin soldier from <Image 1>. Preserve her exact face, glasses, clothing, hand position, toy design, workbench, and lighting. She finishes painting the tin soldier and places it upright on the table. The soldier suddenly salutes her. She leans closer in disbelief. The tiny soldier then marches confidently across the workbench while she follows it with the paintbrush still in her hand. The camera begins with the same clear medium shot, then lowers slightly toward table level to follow the miniature soldier. Playful magical comedy, realistic miniature movement, warm…

Try this prompt →
Prompt

Use the woman and tin soldier from <Image 1>. Preserve her exact face, glasses, clothing, hand position, toy design, workbench, and lighting. She finishes painting the tin soldier and places it upright on the table. The soldier suddenly salutes her. She leans closer in disbelief. The tiny soldier then marches confidently across the workbench while she follows it with the paintbrush still in her hand. The camera begins with the same clear medium shot, then lowers slightly toward table level to follow the miniature soldier. Playful magical comedy, realistic miniature movement, warm…

Try this prompt →
Prompt

The woman in @Image 1 stands at the counter of @Image 2, hands over a coffee with a warm smile and says: "Here you go, enjoy." Medium shot, bright daylight.

Try this prompt →

Specs

Typereference-to-video
OutputVideo
Input-
Model IDbytedance/seedance-2.0-fast/reference-to-video

Similar models

All models →

Prices are per render and charged from your balance only when a render succeeds. Failed renders are refunded automatically. Examples show what the model can do; your result depends on your prompt and settings.