First-person male perspective. On the living room sofa, a young woman (facial reference @【@image-1】) is lying on her side, wearing loose white pajamas, one hand supporting her head, the other holding a phone scrolling…
All Seedance Prompts
Explore prompts for character consistency, brand assets, and AI-powered creative tools.
TOTAL: 816
The main character is [@image-1]. The scene is a breakfast stall on a city street in the morning, with the stall referencing [@image-2]. The camera angle is fixed, referencing [@image-1]. The kitten uses its left hand…
[A tokusatsu video where a giant @ Image1 and a giant @ Image2 fight in a mountainous area. @ Image1 closes the distance and delivers a punch and then a kick to @ Image2. @ Image2 guards and counterattacks with a beam.…
Style: ultra-cinematic photoreal sci-fi, dense futuristic megacity, noir cyberpunk rain, high contrast, metallic reflections, neon accents, wet surfaces, volumetric rain haze, subtle film grain Core motion: ONE…
360-degree panoramic camera selfie. The camera rotates counter-clockwise, filming the interior of a sweets shop. Following this, a woman is shown posing in various scenes, changing her costume and props.
15-second vertical screen 9:16, film grain filter + natural grain, Tamsui River bank walkway in Taipei, background golden sunset + sparkling river surface + silhouette of the Lover's Bridge in the distance, protagonist…
A man walks out of his house towards the parked car. [Cut] The man gets into the car. [Cut] The man starts the engine; the flower on the dashboard trembles slightly. [Cut] The man accelerates quickly and drives wildly.…
Race sequence, presented in rapid cinematic coverage with dynamic angles and intense motion energy - QUICK CUTS - BEAT SYNC HARD EDIT FORMAT: 15s / 120 BPM / 12 SHOTS / HARD CUT.
[Subject]: Wuhan city portrait, including the red walls and yellow tiles of Yellow Crane Tower, steaming hot dry noodles, the steel structure of the Yangtze River Bridge, and cherry blossoms fluttering at Wuhan…
The two female warriors engage in an ultra dynamic sword fight on the neon-lit rooftop. Intense sword clashing with sparks flying, agile dodges, and powerful strikes. Cinematic camera movement tracking the action, high…
The camera overlooks from a high angle. The woman unconsciously lightly strokes the pillow with fingers gripping the edge of the sheet, then slowly raises her face with sleepy eyes, gazing vaguely at the camera from…
A vibrant modern cityscape of Zibo, interwoven with the magnificent landmark Haidai Tower night view and the antique architecture of Zhoucun Ancient Commercial City. The screen contains extremely rich and rapid scene…
[Subject]: Guangzhou city portrait, including ancient Xiguan mansions, exquisite Cantonese dim sum, and the majestic Canton Tower (Little Waist), with rich and saturated colors. [Action]: The camera slowly pushes in…
**### Visual Style** 9:16 vertical composition, 4K ultra-clear realistic quality. Low-key lighting style, high contrast. The cool-toned environmental background forms a sharp contrast with the subject's all-black…
"movie scene, people walking down the streets of the Bronx in 1980, when suddenly they turn into zombies and run like crazy. Funny movie"
High-speed follow-up from a first-person perspective: Closely following a white-haired 3D anime style girl from behind, she is skateboarding at extreme speed down a steep and winding mountain road. The image has a…
Section 1: 16:9 landscape, traditional Chinese paper-cut animation style, with vermilion, golden yellow, and indigo blue as the main colors, clear cutout textures, and obvious paper texture. 0-3 seconds: Long shot,…
0-3 seconds: Chaos begins, jade essence surges. The screen starts completely black, without any light source. As the low sound of a jade chime and ice cracking effects begin, an extremely faint cyan-blue particle stream…
How to use this prompt library
This page is meant to be a reusable Seedance prompt library, not a pile of isolated cards. Start with an example close to your target output, then compare how subject, camera, motion, and references are written before you rewrite anything.
1. Start from the closest output
Pick the example nearest to the visual result you want and inspect which subject, camera, and mood phrases are carrying the outcome.
2. Change one variable at a time
When iterating, adjust lighting, motion, emotion, or references one by one so you can see what actually changed.
3. Use guides and glossary for depth
If the prompt card is not enough, move into the guides for workflow and the glossary for control language.
Keep exploring Seedance content
After the prompt library, use guides for workflow depth or the glossary to decode the terms behind the prompts.
Guides
Read guides for full workflows
Move from examples to method and understand why the prompt is written that way.
Browse
Glossary
Clarify terms before narrowing prompts
Resolve camera, reference, and control language before the next iteration.
Browse
Home
Return home to switch content hubs
Use the homepage to jump between Prompt Library, Guides, and Glossary.
Browse