Girl Skating Power Lines Restores a Town's Lights
A 15-second Japanese full-color 2D anime sequence in which a bob-haired schoolgirl glides along power lines above a shuttered shopping street, jumping between wires as night fog swallows the town and its lights go out one by one. Each glide reignites the street lamps, and she leaps gaps using trails of light as her only foothold, ending on a high perch looking down at the restored town. The prompt specifies high-density, film-quality animation with varied compositions, camera angles, and character distances, no subtitles, watermark, or BGM, but with ambient sound.
Clip is hotlinked from X — if the original post is removed, this clip may stop playing.
Clip specifications
| Model | Seedance 2.0 — ByteDance |
|---|---|
| Duration | 15s |
| Aspect ratio | 1:1 |
| Camera | Not specified |
| Lighting | Not specified |
| Mood | cinematic · hopeful |
| Difficulty | intermediate |
| Prompt language | Chinese |
| Creator | 妖精アーヤさん (@aiehon_aya) |
What makes this clip work
- MOOD
- "cinematic, hopeful" — mood and aesthetic tone are explicitly directed so the emotional register stays consistent from the first frame to the last.
- DURATION
- At 15s this clip is sized for high temporal coherence — long enough for a motion beat, short enough to stay loopable.
- MODEL
- Seedance 2.0 (ByteDance) provides the underlying diffusion and motion architecture. Camera, lighting and duration cues transfer across models — what differs is how literally each model interprets motion.
- PROMPT LANGUAGE
- The prompt is written in Chinese. Every prompt is kept in its original language rather than translated, because the exact words are what was conditioned during generation.
Benchmark methodology is documented in our prompt statistics breakdown.
How to reuse this clip's prompt▾
- Open the full prompt page — the verbatim prompt text lives on the companion prompt page for this clip.
- Copy the text unchanged. Don't paraphrase: the phrasing here is what produced this exact motion and lighting.
- Paste into Seedance 2.0 with aspect ratio 1:1 and clip length 15s, then generate.
- Remix one variable at a time — swap the camera move or the lighting word, keep the rest, and compare the two renders side by side.
Tell bigger stories with Seedance 2.5
Create AI videos up to 30 seconds long with richer image, video, and audio references. Ideal for complex ideas that need consistency and control.
