Two Anime Girls Find a Sleepy Cat on a Flower Street

A 15-second 2D anime short in which two friends discover a napping ginger cat on a colorful Japanese street, and its yawn spreads to one of them. The prompt uses three reference images for character and setting consistency, with shot-by-shot timing, Japanese dialogue, and lip sync.

This clip could not be loaded — it may have been removed from X.
とすくん @tokyo_Valentine ·
View on X ↗

Prompt

MiniMax H3 MAX Prompt
subject_definitions:
<Subject 1>: The girl in Image 1, with a white-blonde bob, blue eyes, colorful rabbit-ear headphones, purple patchwork clothing, white gloves, and decorated sneakers.
<Subject 2>: The girl in Image 2, with long black hair, blunt bangs, blue eyes, rabbit-ear headgear, turquoise clothing, a colorful samurai-inspired skirt, and chunky boots.
<Subject 3>: The street in Image 3, with yellow and red storefronts, potted flowers, flower-painted pavement, overhead wires, and a bright blue sky.
summary:
[reference generation] A 15-second, 16:9 cozy Japanese 2D anime short. Two friends find a sleepy ginger cat in <Subject 3>. Its yawn becomes contagious. Three gentle shots with Japanese dialogue.
retention_analysis:
Preserve both girls' distinct faces, proportions, hairstyles, outfits, and accessories, and the street's colors and floral details. Each character sheet represents one girl. Show exactly two girls in one shared scene. Their swords remain fully sheathed.
detailed_description:
Clean hand-drawn linework, soft cel shading, painted backgrounds, warm afternoon sunlight, fluid animation, and relaxed expressions. Natural real-time motion and quiet pauses. Maintain character positions across cuts.

[Shot 1]
Eye-level medium-wide shot in <Subject 3>. <Subject 1> walks on the left, <Subject 2> on the right. The camera tracks backward slowly. Hair, sleeves, and ornaments sway gently. A ginger cat sleeps beside a blue flowerpot on the right side of the walkway.
<Subject 1> notices it and stops, her grin softening. <Subject 1> (S1) says in a bright, gentle voice, [Japanese] あ、ねこ。 <Subject 2> stops and follows her gaze. Footsteps fade into quiet ambience.
[Shot 2] At 00:05.000, cut to a steady low medium shot. Both girls crouch beside the flowerpot, keeping their left-right positions, hands resting on their knees. The cat is in the foreground, both faces visible above it.
<Subject 2> (S2) tilts her head and says softly, [Japanese] 気持ちよさそう。 She smiles. The cat slowly opens its eyes, stretches its front paws, and gives one lazy yawn. Both girls watch. Nearby flowers stir in the breeze.

[Shot 3] At 00:10.000, cut to a closer two-shot with a subtle, slow push-in. <Subject 1> blinks sleepily, lifts one gloved hand toward her mouth, and yawns. After the yawn, <Subject 1> (S1) says drowsily, [Japanese] 私も、ねむくなっちゃった。 <Subject 2> gives her an affectionate smile. <Subject 1> returns a bashful grin. Hold their shared smile through the final second as the cat settles down.
Use natural Japanese lip sync and separate speaking turns. Full-screen animation without sheet panels, captions, or overlays.

overall_soundscape:
Soft footsteps, fabric rustling, distant birds, and a light breeze. Clear, quiet dialogue.
non_diegetic_music:
Slow, soft piano with sparse glockenspiel notes, kept beneath dialogue.
? How to use this prompt ▾
  1. Copy the prompt using the green "Copy prompt" button above.
  2. Open your model — launch Seedance 2.0 on its official interface.
  3. Paste & generate — keep the verbatim prompt text for best adherence. Set aspect ratio to 16:9 and clip duration to 15s.

What makes this prompt work

Every insight below is verified against benchmark generation parameters, not asserted. How our prompt engineering benchmarks work →

CAMERA
"tilt" is a core cinematic camera technique for establishing spatial immersion and subject framing. Also present: "tracking", "wide shot", "medium shot".
LIGHTING
"natural light, warm tone" establishes luminance depth, volumetric shadow falloff, and tonal contrast across the generation.
MOOD
"cozy" — Mood and aesthetic tone are explicitly directed to maintain emotional consistency throughout the scene.
DURATION
At 15s, this clip is optimized for high temporal coherence and fluid, realistic motion physics.
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. This library keeps every prompt in its original language rather than translating it — the exact words are what was conditioned during generation.
Seedance 2.5 BytePlus Cloud Partner

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.

Explore on BytePlus Official ByteDance Enterprise Cloud