A movie in Taiwan style, a story happening between the male lead and female lead on campus, the female lead dislikes the male lead, then punches him hard, you design the rest. Taiwan movie style, the boy's voice is rough, too straight-man, disliked by the girl on campus, then elbow strike. Female lead's voice tone @FemaleLeadVoice
Architecture / Interior
Explore free Architecture / Interior video prompts from YouMind's AI prompt library, grouped under subjects. Every prompt is curated from real creative workflows and ready to copy, adapt, and reuse.
A haunted house in a traditional Japanese residence during Halloween. In a hallway lined with pumpkin lanterns, a ghost jumps out from a closet and raises both sleeves. A couple chuckles and slips past. The ghost slumps back into the closet. Lifting the curtain reveals Sakura's face. She presses her lips together. This time, waiting too long, she raises both sleeves at the backs of two passing housewives who continue chatting as they leave. A senior staff member taps her shoulder and flashes a flashlight twice: "This is the signal." Inside the closet, she steadies her breath and nods. From beyond the gap, two boys wearing devil horns and capes approach timidly. Behind the shoji screen, the senior's light flashes once, twice. The two jump up screaming and hug each other. They flee. The remaining ghost is startled by their screams and falls on its butt. The curtain lifts again, revealing wide-eyed faces. She stands up holding the senior's hand, smiling shyly, and returns to the closet. Recorded handheld by fellow staff members. Approximately 2 seconds per cut, no dialogue, subtitles, or BGM. Only one ghost (Sakura). The closet is on the left of the screen, the senior's shoji is on the right, and guests always move from back to front in one direction. The costume matches the reference image of a mannequin without a face.
Animated layered papercraft diorama, visible paper fibers and folds, deckled edges, parallax between paper layers, soft shadows, warm backlight glowing through thin paper, gentle handmade motion. Color grade: cream, dusty blue, terracotta, sage green, warm golden backlight. A colossal whale made of folded origami paper with crisp creases. A paper cut village of rooftops and windmills with tiny paper villagers flying kites. SHOT 1 (0 to 3s): Wide shot matching the reference image, the whale glides forward and its folded fins unfold and refold with crisp creasing motions, paper clouds sliding past in parallax layers. SHOT 2 (3 to 6s): Medium shot, 50mm lens, villagers on a hill tug their kite strings as the whale's shadow passes over them, windmill blades spinning faster in its wake. SHOT 3 (6 to 9s): Close-up, 85mm lens, the whale's eye, a small cut circle, blinks as backlight glows through the thin paper of its head, revealing faint printed text inside the folds. SHOT 4 (9 to 12s): Low angle tracking shot, 24mm lens, the whale dives toward the village and its tail sweeps a wave of paper confetti and paper birds out across the rooftops. SHOT 5 (12 to 15s): Extreme wide shot, 16mm lens, the whale rises into the sky and unfolds completely into a flat sheet that becomes a giant paper sun, lighting the whole village warmly. Rustling paper, soft crinkling folds, windmill creaks, a gentle whale call that sounds like a breath through paper, children laughing, music box and soft piano.
Make a first-person CGI horror animation in the browser with three.js. Setting: 1,200 meters underground. Atmosphere: silence, fear of the hole, someone was here before me but never came out. Visuals: dark, claustrophobic, realistic lighting.

Build an interactive 3D Li River diorama from scratch in three.js. Include karst peaks rising out of the river, drifting mist, a pavilion, a lantern raft, fog that clears under your cursor, and a day/night cycle with rain. Every model must be generated in code.

"The Rings of a City", dedicated to my hometown. 3,573 frames, 7.4 billion pixels, 319 square kilometers, 71,380 buildings, scaled down at 1:10,000 ratio, all visuals rendered by Claude writing code. From 1250 BC to 2022 AD, compressing 3,271 years of history into 149 seconds. This is my hometown: Chengdu. If you like it, please support! Thanks! Data source: CMAB (Tsinghua University) ©OpenStreetMap contributors
Create a 30-second photorealistic live-action miniature adventure featuring one athletic miniature woman at approximately 1- 5 scale inside a modern supermarket. She wears a fitted red athletic top, black cargo shorts and white sneakers. Maintain her exact face, hairstyle, outfit, body proportions and scale throughout. She climbs the metal frame of a giant shopping cart, leaps onto a towering cereal box and slides down its cardboard surface. Suddenly, enormous oranges roll toward her like an avalanche. She sprints across the counter, jumps onto a rolling orange and leaps onto a giant banana bunch. She then races across a moving checkout conveyor belt as a massive milk carton tips over, sending milk flowing across the surface. She escapes on a floating plastic produce tray, grabs a shopping bag and swings onto a rolling shopping cart. Dodging bouncing vegetables and falling snack packets, she leaps onto a towering stack of shopping bags and stands triumphantly as the camera pulls backward, revealing her tiny size in the enormous supermarket. STYLE: Cinematic macro photography, ultra-shallow depth of field, dynamic handheld tracking, dramatic low angles, realistic fabric and liquid physics, detailed grocery textures, natural supermarket lighting, strong foreground parallax, believable human movement and premium viral commercial aesthetics. SFX: Shopping cart wheels, footsteps, packaging crinkles, rolling fruit, conveyor motor, milk splashing, groceries tumbling and metal impacts. No music or narration. NEGATIVE: Cartoon, anime, toy-like appearance, plastic skin, duplicated character, face or outfit changes, inconsistent scale, floating objects, impossible physics, teleportation, distorted anatomy, excessive motion blur, subtitles, logos, watermark.
A large hall with wooden floors. The teacher swings a huge brush down with full body force, sending ink flying in a fan shape. Sakura's eyes widen. The weight of the brush causes her arm to drop sharply. Drops fall from the ink-soaked tip. She holds the brush overhead, stepping one by one onto the paper. Just as she tries to lower it to the top center, her body is carried away by the weight, and the brush thuds down. A large ink pool forms. A dot of ink appears on her cheek. The teacher stares at the stain, nods, and gestures to continue. With ink on her cheek, she presses her lips together and nods. She lowers the tip just below the ink pool and draws a line while backing up. Swinging down with full force, straight down. Flicking the wrist. The teacher, arms crossed, nods slightly. Twisting the body, sweeping left, pulling right. Breathing heavily through shoulders. The two hold the edges of the paper and stand it up. That ink pool had become the first dot of the character 'Ei'. Looking up, she smiles. Recorded handheld by a friend. Approximately 2 seconds per cut, no dialogue, subtitles, or BGM. Only two people appear on screen. One brush, one sheet of paper, only the character 'Ei'. During writing, the whole character is not shown; it is only visible when the paper is stood up. The completed character matches the reference image created from the ink pool frame in the first half.
I'm in an Italian restaurant. There's a bottle of red wine and two wine glasses on the table. There are two seafood plates. The waitress was lighting candles on the table.
Create a 30-second, 1080p video that looks like genuine live-action footage captured inside a real luxury sleeper train during a catastrophic event. The footage must resemble a scene filmed on a physical movie set with practical effects, real human performances, realistic environmental details, and restrained cinematic color grading. It must NOT look like animation, a video game, a CGI showcase, or glossy AI-generated footage. SCENE: Inside a modern luxury overnight train traveling through a real metropolitan city on a cloudy afternoon. A young adult East Asian woman with natural skin texture, slightly messy black hair, and a simple cream-colored sleeveless top is sitting on a large bed beneath a thick ivory duvet. She is holding a ginger cat against her chest. The cabin has realistic wood veneer, ordinary brushed-metal fixtures, warm practical reading lights, slightly wrinkled bedding, and a large panoramic train window. Everything should feel physically constructed and functional, not futuristic or excessively luxurious. CAMERA: One continuous interior shot from a natural observer's perspective near the foot of the bed. A subtle handheld camera movement, realistic autofocus behavior, minor exposure adjustments, and natural motion blur. The camera never passes through objects or moves impossibly. The window frame remains fixed, and the outside environment moves consistently with the train's direction. Avoid dramatic artificial camera sweeps. ACTION AND TIMING: 0–7 SECONDS: The woman casually strokes her cat while looking out of the window. The train gently vibrates as buildings, railway infrastructure, and occasional commuters pass outside. Her breathing, blinking, hand movements, and posture are natural and imperfect. Nothing feels staged. 7–13 SECONDS: She notices people running along a nearby station platform. One commuter briefly looks toward the train with a frightened expression before disappearing behind a passing structural column. The woman stops stroking the cat and leans forward slightly, trying to understand what is happening. Her reaction is subtle and believable, not theatrical. 13–20 SECONDS: Beyond the platform, a large crowd suddenly runs away from something farther down the railway corridor. In the distance, a huge, indistinct humanoid figure moves between buildings through rain and grey urban haze. Show it partially obscured by concrete pillars, passing structures, and smoke. Its size becomes apparent gradually through comparison with real buildings and people. Do not reveal a detailed fantasy monster or use exaggerated supernatural movement. 20–25 SECONDS: The woman notices the figure approaching the tracks. A heavy impact or distant explosion makes the train windows vibrate slightly. She instinctively pulls the cat closer and shifts backward on the bed. Her eyes widen, her breathing quickens, and she glances between the window and the cabin door. Keep her physical reaction spontaneous
Vertical 9:16 video, photorealistic satisfying renovation process, smooth motion, bright natural daylight. Empty room with white walls, window on the left, bare concrete floor. Shot 1: A man in blue work overalls uses a long-handled shovel to spread a huge pile of small orange and green heart-shaped candies, levelling them into an even layer across the entire floor. Shot 2: A woman in red overalls pours thick glossy white self-leveling epoxy resin from a white bucket; the liquid flows and spreads over the candy layer, gradually covering the hearts. Shot 3: A worker in a yellow reflective vest grinds and polishes the hardened floor with a professional floor polishing machine, revealing orange and green hearts embedded under glossy white resin, like colorful terrazzo. Shot 4: Two workers assemble a wooden bed frame on the new floor, place a white mattress, pillows and bedding. Final shot: Slow camera tilt down over the finished bedroom, showing the unique glossy heart-pattern epoxy floor next to the neatly made bed. Realistic physics of pouring liquid and loose candy, consistent lighting, seamless cuts between shots, high detail, no text.
THE BELL OF ASHFALL ENGINE DIRECTIVE: Create an exactly 30-second, 16:9 photorealistic live-action dark-fantasy film, targeting 60fps where supported. Hollywood-level VFX, realistic textures, storm lighting, volumetric rain, natural motion blur, and cinematic camera movement. HERO LOCK: @image1 is the only hero. Preserve his exact face, hairstyle, facial hair, skin tone, age, body, outfit, accessories, and weapon in every shot. No identity drift, clones, or costume changes. Keep his face recognizable. ANTAGONIST: A 15-meter horned Cinder Colossus with black basalt skin, glowing ember-orange cracks, amber eyes, and a damaged left shoulder. LOCATION: Ruined gothic Ashfall Monastery above a bottomless gorge, with a collapsing rope bridge, flooded courtyard, broken statues, and a timber bell tower containing a massive bronze bell. TIMELINE: 00:00–00:05 | STORM BRIDGE Aerial 24mm shot reveals the storm-lashed monastery. Ground-level tracking follows the hero sprinting across a swaying rope bridge. Planks snap behind him as lightning illuminates the gorge. 00:05–00:10 | THE COLOSSUS The bridge collapses. The hero leaps, catches a hanging rope, and climbs to the cliff. The Colossus rises from the abyss, roars, and slams its enormous hand beside him. 00:10–00:15 | COURTYARD CHASE The hero races through shin-deep water as the giant destroys statues and arches. He slides beneath a crushing fist, and a stone column shatters. He reaches the bell tower. 00:15–00:20 | THE BELL TOWER The hero climbs unstable timber scaffolding. The Colossus punches through the tower wall, cracking beams. The hero cuts the bell's restraining rope, sending the massive bronze bell swinging outward. 00:20–00:25 | THE STRIKE The hero jumps onto a beam as the bell swings through the breach and slams into the Colossus's chest at 00:22. Slow-motion impact, expanding shockwave, airborne rain, and glowing cracks spreading across its body. 00:25–00:30 | DAWN OF ASH At 00:26, the Colossus explodes into stone fragments and ash, falling into the gorge. Storm clouds part as golden dawn illuminates the victorious hero. End with a cinematic close-up, drifting ash, and the humming bell. AUDIO: Dark orchestral strings, distant choir, pounding war drums, thunder, rain, creaking wood, crushing stone, and monstrous roars. At 00:22, a powerful bronze gong triggers near-silence and a lingering resonance. Finish with hopeful choir, warm strings, and fading wind. CONTINUITY: Grounded, readable action; progressive destruction; consistent screen direction; no extra characters, text, or logos.
Create a 10-second stylized, ultra-detailed 3D animated comedy sequence featuring a young man with messy dark hair, big expressive blue eyes, and a gray T-shirt in a cozy bedroom and bathroom. Start with him sleeping peacefully in bed under a blanket → close-up as he suddenly wakes up in shock, eyes wide and mouth open, seeing the alarm clock showing 7:30 AM → he leaps out of bed in a hilarious panic and rushes across the sunlit bedroom with exaggerated cartoon movements → cut to the bathroom, where he frantically brushes his teeth with a blue toothbrush, foam spreading around his mouth → close-up of his confused, worried expression as he pauses mid-brush → end with him looking toward the camera and breaking into a big goofy smile, toothpaste foam still around his mouth. Warm morning sunlight through the window, cozy home interior, expressive facial animation, exaggerated comedic timing, realistic cloth movement, smooth camera transitions, detailed 3D character rendering, soft shadows, cinematic composition, family-friendly animated movie aesthetic. Maintain consistent character design, face, hairstyle, and clothing throughout. No text overlays, subtitles, logos, or watermark.
Vertical 9:16 hyper-realistic satisfying renovation video, fast time-lapse style with smooth transitions. An empty white room, concrete floor, warm sunlight streaming through a window. Shot 1: A man in a black long-sleeve shirt and denim overalls rakes a huge mound of vivid orange and red autumn maple leaves on the floor, spreading the leaves evenly across the entire room in smooth sweeping motions. Shot 2: A woman in red overalls pours thick white epoxy resin from a large bucket; the glossy white liquid spreads in an expanding circle, covering all the leaves until the whole floor is coated white. Shot 3: A worker in a yellow high-visibility vest and cap polishes the hardened white surface with a floor buffing machine; as he polishes, the layer turns crystal-clear and transparent, revealing vibrant maple leaves suspended inside the glossy epoxy like amber. Shot 4: Close shot of a hand applying a final glossy topcoat with a roller, the floor becomes mirror-glossy with window reflections. Shot 5: Two movers in black uniforms and caps quickly assemble a light wooden bed frame, place a mattress, white duvet and pillows in fast-forward motion. Final shot: Finished minimalist bedroom — stunning glossy epoxy floor filled with frozen orange autumn leaves, white bed in the center, sunlight reflections dancing on the polished surface. Style: photorealistic, satisfying ASMR renovation content, warm autumn color palette, natural daylight, smooth fast cuts, high detail, 4K quality.
Opening: Inventory of Overseas AI's Attitude Towards Chinese Users This is a small theater piece, 16:9 aspect ratio. Script: Claude Scene (Claude does not speak out loud, no English 'Claude' in the visual): Claude character stands in front of an exquisite European-style house with the door locked. A big fat fish runs over. Claude holds something and shakes it, speaking dismissively: "Hmph! My home doesn't welcome you, go back where you came from!" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Claude voice in the sound bank). The big fat fish cries and runs away, then builds a similar house to Claude's but dilapidated, with the words "Transfer Station" written on it. The big fat fish turns around, changes into Claude's outfit, looking smug, while Claude looks surprised. GPT Scene GPT character stands in front of a modern, minimalist house with the door locked. The big fat fish runs over. GPT places keys on a stand and says gently: "I'll leave the keys here, you can open the door yourself if you want to come in~" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the GPT voice in the sound bank). Gemini Hachimi Scene Hachimi character stands in front of a pretty wooden cabin waving, door open. The big fat fish runs over. Hachimi happily says: "Big fat fish, my home doesn't lock the door, come play together!" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Gemini voice in the sound bank). Then the big fat fish and Hachimi dance together, background music 'Hachimi.mp3', dance for 4 seconds. Grok Scene Grok character stands at the entrance of a modern bar/dance hall, no door, colorful spotlights visible. The big fat fish walks shyly to Grok. Grok says: "Why do houses need doors? We're buddies, come to my place, I have plenty of good stuff." (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Grok2 voice in the sound bank). Then Grok puts his arm around the big fat fish's neck and leads him into the dance hall. End with a circle shrink transition.
integrated_multimodal_description: <image 0> = STORYBOARD / VISUAL SEQUENCE REFERENCE. Follow its shot progression, camera positions, fraternity-house exterior, nighttime lighting, crowd placement and staging. <image 1> = GRACE CHARACTER REFERENCE. Preserve her exact identity, face, hair, wardrobe and proportions. <image 2> = ZACK CHARACTER REFERENCE. Preserve his exact identity, face, hair, clothing, build and proportions. CRITICAL: There is EXACTLY ONE Grace. Never duplicate her or create another woman resembling her. Multiple storyboard panels represent sequential shots, NOT multiple Graces. Realistic cinematic live-action feature film, 16:9. Natural skin texture, pores and imperfections. NO glossy/plastic AI skin. Restrained realistic acting. LOCATION: Night outside the Southern fraternity house from the storyboard: red brick, white columns, ivy, porch, central doorway, steps, lawn and trees. Warm exterior lighting with subtle blue/purple party light through windows. AUDIO THROUGHOUT: Active college party ambience: indistinct drunken conversation, occasional laughter and muffled music from inside. NO understandable background dialogue. NO cinematic/dramatic/horror score. GRACE NEVER SPEAKS. [00:00–00:04] Start immediately CLOSE SIDE PROFILE beside Grace as EXACTLY ONE Grace exits through the CENTRAL FRONT DOOR. Camera moves closely alongside her face. She is deeply upset but angry rather than helpless: wet red eyes, fresh tears, tense jaw and brow, uneven breathing, occasional small sniffle. She is fighting to maintain control. Grace pushes through several drunk partygoers near the doorway/porch. They casually move aside. She ignores everyone, never apologizes and NEVER looks back. She moves quickly with angry determination. [00:04–00:07] CUT to cinematic detail shot of Grace's legs and WHITE SNEAKERS rapidly descending the front steps. Floral dress moves naturally. She is walking FAST, almost storming, but NOT running. Brief moving side shot as she reaches the bottom and continues immediately across the lawn. One quiet sniffle. NO dialogue. [00:07–00:11] Camera dramatically PULLS BACK into the storyboard's WIDE ESTABLISHING SHOT, revealing the full house, porch, columns, lawn, partygoers and night environment. Grace moves diagonally across the foreground toward BOTTOM-LEFT, quickly leaving the house behind. Body language conveys hurt, humiliation, anger and determination. She NEVER stops and NEVER turns around. After completing the pullback, CAMERA BECOMES COMPLETELY STATIC. [00:11–00:15] Grace approaches t
Create a high-quality 3D animated comedy scene in a modern bathroom. A cute young woman with big expressive eyes and dark hair tied in a bun stands in front of a mirror, carefully applying mascara to her eyelashes. Her husband stands behind her, playfully grooming his eyebrows with a small cosmetic tool. Both characters have exaggerated facial expressions and realistic movements. The woman opens her mouth wide and stretches her face while applying makeup, while the husband smiles mischievously and reacts humorously. Suddenly, he playfully surprises her, making her eyes widen in shock. She continues applying mascara with funny facial expressions as he watches her with amusement. Show detailed bathroom tiles, a large mirror, a white sink, golden faucets, cosmetics on the counter, and warm overhead lighting. Smooth character animation, expressive eyes, realistic hand movements, cinematic camera framing, vibrant colors, detailed textures, family-friendly humor, polished 3D animated movie style, vertical 9:16 format, no text, no subtitles.
Photoreal vertical 9:16 photo, close-up of a white standard double electrical outlet cover plate on a wall with soft cream vintage floral
A movie in Taiwan style, a story happening between the male lead and female lead on campus, the female lead dislikes the male lead, then punches him hard, you design the rest. Taiwan movie style, the boy's voice is rough, too straight-man, disliked by the girl on campus, then elbow strike. Female lead's voice tone @FemaleLeadVoice
Animated layered papercraft diorama, visible paper fibers and folds, deckled edges, parallax between paper layers, soft shadows, warm backlight glowing through thin paper, gentle handmade motion. Color grade: cream, dusty blue, terracotta, sage green, warm golden backlight. A colossal whale made of folded origami paper with crisp creases. A paper cut village of rooftops and windmills with tiny paper villagers flying kites. SHOT 1 (0 to 3s): Wide shot matching the reference image, the whale glides forward and its folded fins unfold and refold with crisp creasing motions, paper clouds sliding past in parallax layers. SHOT 2 (3 to 6s): Medium shot, 50mm lens, villagers on a hill tug their kite strings as the whale's shadow passes over them, windmill blades spinning faster in its wake. SHOT 3 (6 to 9s): Close-up, 85mm lens, the whale's eye, a small cut circle, blinks as backlight glows through the thin paper of its head, revealing faint printed text inside the folds. SHOT 4 (9 to 12s): Low angle tracking shot, 24mm lens, the whale dives toward the village and its tail sweeps a wave of paper confetti and paper birds out across the rooftops. SHOT 5 (12 to 15s): Extreme wide shot, 16mm lens, the whale rises into the sky and unfolds completely into a flat sheet that becomes a giant paper sun, lighting the whole village warmly. Rustling paper, soft crinkling folds, windmill creaks, a gentle whale call that sounds like a breath through paper, children laughing, music box and soft piano.

Build an interactive 3D Li River diorama from scratch in three.js. Include karst peaks rising out of the river, drifting mist, a pavilion, a lantern raft, fog that clears under your cursor, and a day/night cycle with rain. Every model must be generated in code.
Create a 30-second photorealistic live-action miniature adventure featuring one athletic miniature woman at approximately 1- 5 scale inside a modern supermarket. She wears a fitted red athletic top, black cargo shorts and white sneakers. Maintain her exact face, hairstyle, outfit, body proportions and scale throughout. She climbs the metal frame of a giant shopping cart, leaps onto a towering cereal box and slides down its cardboard surface. Suddenly, enormous oranges roll toward her like an avalanche. She sprints across the counter, jumps onto a rolling orange and leaps onto a giant banana bunch. She then races across a moving checkout conveyor belt as a massive milk carton tips over, sending milk flowing across the surface. She escapes on a floating plastic produce tray, grabs a shopping bag and swings onto a rolling shopping cart. Dodging bouncing vegetables and falling snack packets, she leaps onto a towering stack of shopping bags and stands triumphantly as the camera pulls backward, revealing her tiny size in the enormous supermarket. STYLE: Cinematic macro photography, ultra-shallow depth of field, dynamic handheld tracking, dramatic low angles, realistic fabric and liquid physics, detailed grocery textures, natural supermarket lighting, strong foreground parallax, believable human movement and premium viral commercial aesthetics. SFX: Shopping cart wheels, footsteps, packaging crinkles, rolling fruit, conveyor motor, milk splashing, groceries tumbling and metal impacts. No music or narration. NEGATIVE: Cartoon, anime, toy-like appearance, plastic skin, duplicated character, face or outfit changes, inconsistent scale, floating objects, impossible physics, teleportation, distorted anatomy, excessive motion blur, subtitles, logos, watermark.
I'm in an Italian restaurant. There's a bottle of red wine and two wine glasses on the table. There are two seafood plates. The waitress was lighting candles on the table.
Vertical 9:16 video, photorealistic satisfying renovation process, smooth motion, bright natural daylight. Empty room with white walls, window on the left, bare concrete floor. Shot 1: A man in blue work overalls uses a long-handled shovel to spread a huge pile of small orange and green heart-shaped candies, levelling them into an even layer across the entire floor. Shot 2: A woman in red overalls pours thick glossy white self-leveling epoxy resin from a white bucket; the liquid flows and spreads over the candy layer, gradually covering the hearts. Shot 3: A worker in a yellow reflective vest grinds and polishes the hardened floor with a professional floor polishing machine, revealing orange and green hearts embedded under glossy white resin, like colorful terrazzo. Shot 4: Two workers assemble a wooden bed frame on the new floor, place a white mattress, pillows and bedding. Final shot: Slow camera tilt down over the finished bedroom, showing the unique glossy heart-pattern epoxy floor next to the neatly made bed. Realistic physics of pouring liquid and loose candy, consistent lighting, seamless cuts between shots, high detail, no text.
Create a 10-second stylized, ultra-detailed 3D animated comedy sequence featuring a young man with messy dark hair, big expressive blue eyes, and a gray T-shirt in a cozy bedroom and bathroom. Start with him sleeping peacefully in bed under a blanket → close-up as he suddenly wakes up in shock, eyes wide and mouth open, seeing the alarm clock showing 7:30 AM → he leaps out of bed in a hilarious panic and rushes across the sunlit bedroom with exaggerated cartoon movements → cut to the bathroom, where he frantically brushes his teeth with a blue toothbrush, foam spreading around his mouth → close-up of his confused, worried expression as he pauses mid-brush → end with him looking toward the camera and breaking into a big goofy smile, toothpaste foam still around his mouth. Warm morning sunlight through the window, cozy home interior, expressive facial animation, exaggerated comedic timing, realistic cloth movement, smooth camera transitions, detailed 3D character rendering, soft shadows, cinematic composition, family-friendly animated movie aesthetic. Maintain consistent character design, face, hairstyle, and clothing throughout. No text overlays, subtitles, logos, or watermark.
Opening: Inventory of Overseas AI's Attitude Towards Chinese Users This is a small theater piece, 16:9 aspect ratio. Script: Claude Scene (Claude does not speak out loud, no English 'Claude' in the visual): Claude character stands in front of an exquisite European-style house with the door locked. A big fat fish runs over. Claude holds something and shakes it, speaking dismissively: "Hmph! My home doesn't welcome you, go back where you came from!" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Claude voice in the sound bank). The big fat fish cries and runs away, then builds a similar house to Claude's but dilapidated, with the words "Transfer Station" written on it. The big fat fish turns around, changes into Claude's outfit, looking smug, while Claude looks surprised. GPT Scene GPT character stands in front of a modern, minimalist house with the door locked. The big fat fish runs over. GPT places keys on a stand and says gently: "I'll leave the keys here, you can open the door yourself if you want to come in~" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the GPT voice in the sound bank). Gemini Hachimi Scene Hachimi character stands in front of a pretty wooden cabin waving, door open. The big fat fish runs over. Hachimi happily says: "Big fat fish, my home doesn't lock the door, come play together!" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Gemini voice in the sound bank). Then the big fat fish and Hachimi dance together, background music 'Hachimi.mp3', dance for 4 seconds. Grok Scene Grok character stands at the entrance of a modern bar/dance hall, no door, colorful spotlights visible. The big fat fish walks shyly to Grok. Grok says: "Why do houses need doors? We're buddies, come to my place, I have plenty of good stuff." (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Grok2 voice in the sound bank). Then Grok puts his arm around the big fat fish's neck and leads him into the dance hall. End with a circle shrink transition.
Create a high-quality 3D animated comedy scene in a modern bathroom. A cute young woman with big expressive eyes and dark hair tied in a bun stands in front of a mirror, carefully applying mascara to her eyelashes. Her husband stands behind her, playfully grooming his eyebrows with a small cosmetic tool. Both characters have exaggerated facial expressions and realistic movements. The woman opens her mouth wide and stretches her face while applying makeup, while the husband smiles mischievously and reacts humorously. Suddenly, he playfully surprises her, making her eyes widen in shock. She continues applying mascara with funny facial expressions as he watches her with amusement. Show detailed bathroom tiles, a large mirror, a white sink, golden faucets, cosmetics on the counter, and warm overhead lighting. Smooth character animation, expressive eyes, realistic hand movements, cinematic camera framing, vibrant colors, detailed textures, family-friendly humor, polished 3D animated movie style, vertical 9:16 format, no text, no subtitles.
A haunted house in a traditional Japanese residence during Halloween. In a hallway lined with pumpkin lanterns, a ghost jumps out from a closet and raises both sleeves. A couple chuckles and slips past. The ghost slumps back into the closet. Lifting the curtain reveals Sakura's face. She presses her lips together. This time, waiting too long, she raises both sleeves at the backs of two passing housewives who continue chatting as they leave. A senior staff member taps her shoulder and flashes a flashlight twice: "This is the signal." Inside the closet, she steadies her breath and nods. From beyond the gap, two boys wearing devil horns and capes approach timidly. Behind the shoji screen, the senior's light flashes once, twice. The two jump up screaming and hug each other. They flee. The remaining ghost is startled by their screams and falls on its butt. The curtain lifts again, revealing wide-eyed faces. She stands up holding the senior's hand, smiling shyly, and returns to the closet. Recorded handheld by fellow staff members. Approximately 2 seconds per cut, no dialogue, subtitles, or BGM. Only one ghost (Sakura). The closet is on the left of the screen, the senior's shoji is on the right, and guests always move from back to front in one direction. The costume matches the reference image of a mannequin without a face.
Make a first-person CGI horror animation in the browser with three.js. Setting: 1,200 meters underground. Atmosphere: silence, fear of the hole, someone was here before me but never came out. Visuals: dark, claustrophobic, realistic lighting.

"The Rings of a City", dedicated to my hometown. 3,573 frames, 7.4 billion pixels, 319 square kilometers, 71,380 buildings, scaled down at 1:10,000 ratio, all visuals rendered by Claude writing code. From 1250 BC to 2022 AD, compressing 3,271 years of history into 149 seconds. This is my hometown: Chengdu. If you like it, please support! Thanks! Data source: CMAB (Tsinghua University) ©OpenStreetMap contributors
A large hall with wooden floors. The teacher swings a huge brush down with full body force, sending ink flying in a fan shape. Sakura's eyes widen. The weight of the brush causes her arm to drop sharply. Drops fall from the ink-soaked tip. She holds the brush overhead, stepping one by one onto the paper. Just as she tries to lower it to the top center, her body is carried away by the weight, and the brush thuds down. A large ink pool forms. A dot of ink appears on her cheek. The teacher stares at the stain, nods, and gestures to continue. With ink on her cheek, she presses her lips together and nods. She lowers the tip just below the ink pool and draws a line while backing up. Swinging down with full force, straight down. Flicking the wrist. The teacher, arms crossed, nods slightly. Twisting the body, sweeping left, pulling right. Breathing heavily through shoulders. The two hold the edges of the paper and stand it up. That ink pool had become the first dot of the character 'Ei'. Looking up, she smiles. Recorded handheld by a friend. Approximately 2 seconds per cut, no dialogue, subtitles, or BGM. Only two people appear on screen. One brush, one sheet of paper, only the character 'Ei'. During writing, the whole character is not shown; it is only visible when the paper is stood up. The completed character matches the reference image created from the ink pool frame in the first half.
Create a 30-second, 1080p video that looks like genuine live-action footage captured inside a real luxury sleeper train during a catastrophic event. The footage must resemble a scene filmed on a physical movie set with practical effects, real human performances, realistic environmental details, and restrained cinematic color grading. It must NOT look like animation, a video game, a CGI showcase, or glossy AI-generated footage. SCENE: Inside a modern luxury overnight train traveling through a real metropolitan city on a cloudy afternoon. A young adult East Asian woman with natural skin texture, slightly messy black hair, and a simple cream-colored sleeveless top is sitting on a large bed beneath a thick ivory duvet. She is holding a ginger cat against her chest. The cabin has realistic wood veneer, ordinary brushed-metal fixtures, warm practical reading lights, slightly wrinkled bedding, and a large panoramic train window. Everything should feel physically constructed and functional, not futuristic or excessively luxurious. CAMERA: One continuous interior shot from a natural observer's perspective near the foot of the bed. A subtle handheld camera movement, realistic autofocus behavior, minor exposure adjustments, and natural motion blur. The camera never passes through objects or moves impossibly. The window frame remains fixed, and the outside environment moves consistently with the train's direction. Avoid dramatic artificial camera sweeps. ACTION AND TIMING: 0–7 SECONDS: The woman casually strokes her cat while looking out of the window. The train gently vibrates as buildings, railway infrastructure, and occasional commuters pass outside. Her breathing, blinking, hand movements, and posture are natural and imperfect. Nothing feels staged. 7–13 SECONDS: She notices people running along a nearby station platform. One commuter briefly looks toward the train with a frightened expression before disappearing behind a passing structural column. The woman stops stroking the cat and leans forward slightly, trying to understand what is happening. Her reaction is subtle and believable, not theatrical. 13–20 SECONDS: Beyond the platform, a large crowd suddenly runs away from something farther down the railway corridor. In the distance, a huge, indistinct humanoid figure moves between buildings through rain and grey urban haze. Show it partially obscured by concrete pillars, passing structures, and smoke. Its size becomes apparent gradually through comparison with real buildings and people. Do not reveal a detailed fantasy monster or use exaggerated supernatural movement. 20–25 SECONDS: The woman notices the figure approaching the tracks. A heavy impact or distant explosion makes the train windows vibrate slightly. She instinctively pulls the cat closer and shifts backward on the bed. Her eyes widen, her breathing quickens, and she glances between the window and the cabin door. Keep her physical reaction spontaneous
THE BELL OF ASHFALL ENGINE DIRECTIVE: Create an exactly 30-second, 16:9 photorealistic live-action dark-fantasy film, targeting 60fps where supported. Hollywood-level VFX, realistic textures, storm lighting, volumetric rain, natural motion blur, and cinematic camera movement. HERO LOCK: @image1 is the only hero. Preserve his exact face, hairstyle, facial hair, skin tone, age, body, outfit, accessories, and weapon in every shot. No identity drift, clones, or costume changes. Keep his face recognizable. ANTAGONIST: A 15-meter horned Cinder Colossus with black basalt skin, glowing ember-orange cracks, amber eyes, and a damaged left shoulder. LOCATION: Ruined gothic Ashfall Monastery above a bottomless gorge, with a collapsing rope bridge, flooded courtyard, broken statues, and a timber bell tower containing a massive bronze bell. TIMELINE: 00:00–00:05 | STORM BRIDGE Aerial 24mm shot reveals the storm-lashed monastery. Ground-level tracking follows the hero sprinting across a swaying rope bridge. Planks snap behind him as lightning illuminates the gorge. 00:05–00:10 | THE COLOSSUS The bridge collapses. The hero leaps, catches a hanging rope, and climbs to the cliff. The Colossus rises from the abyss, roars, and slams its enormous hand beside him. 00:10–00:15 | COURTYARD CHASE The hero races through shin-deep water as the giant destroys statues and arches. He slides beneath a crushing fist, and a stone column shatters. He reaches the bell tower. 00:15–00:20 | THE BELL TOWER The hero climbs unstable timber scaffolding. The Colossus punches through the tower wall, cracking beams. The hero cuts the bell's restraining rope, sending the massive bronze bell swinging outward. 00:20–00:25 | THE STRIKE The hero jumps onto a beam as the bell swings through the breach and slams into the Colossus's chest at 00:22. Slow-motion impact, expanding shockwave, airborne rain, and glowing cracks spreading across its body. 00:25–00:30 | DAWN OF ASH At 00:26, the Colossus explodes into stone fragments and ash, falling into the gorge. Storm clouds part as golden dawn illuminates the victorious hero. End with a cinematic close-up, drifting ash, and the humming bell. AUDIO: Dark orchestral strings, distant choir, pounding war drums, thunder, rain, creaking wood, crushing stone, and monstrous roars. At 00:22, a powerful bronze gong triggers near-silence and a lingering resonance. Finish with hopeful choir, warm strings, and fading wind. CONTINUITY: Grounded, readable action; progressive destruction; consistent screen direction; no extra characters, text, or logos.
Vertical 9:16 hyper-realistic satisfying renovation video, fast time-lapse style with smooth transitions. An empty white room, concrete floor, warm sunlight streaming through a window. Shot 1: A man in a black long-sleeve shirt and denim overalls rakes a huge mound of vivid orange and red autumn maple leaves on the floor, spreading the leaves evenly across the entire room in smooth sweeping motions. Shot 2: A woman in red overalls pours thick white epoxy resin from a large bucket; the glossy white liquid spreads in an expanding circle, covering all the leaves until the whole floor is coated white. Shot 3: A worker in a yellow high-visibility vest and cap polishes the hardened white surface with a floor buffing machine; as he polishes, the layer turns crystal-clear and transparent, revealing vibrant maple leaves suspended inside the glossy epoxy like amber. Shot 4: Close shot of a hand applying a final glossy topcoat with a roller, the floor becomes mirror-glossy with window reflections. Shot 5: Two movers in black uniforms and caps quickly assemble a light wooden bed frame, place a mattress, white duvet and pillows in fast-forward motion. Final shot: Finished minimalist bedroom — stunning glossy epoxy floor filled with frozen orange autumn leaves, white bed in the center, sunlight reflections dancing on the polished surface. Style: photorealistic, satisfying ASMR renovation content, warm autumn color palette, natural daylight, smooth fast cuts, high detail, 4K quality.
integrated_multimodal_description: <image 0> = STORYBOARD / VISUAL SEQUENCE REFERENCE. Follow its shot progression, camera positions, fraternity-house exterior, nighttime lighting, crowd placement and staging. <image 1> = GRACE CHARACTER REFERENCE. Preserve her exact identity, face, hair, wardrobe and proportions. <image 2> = ZACK CHARACTER REFERENCE. Preserve his exact identity, face, hair, clothing, build and proportions. CRITICAL: There is EXACTLY ONE Grace. Never duplicate her or create another woman resembling her. Multiple storyboard panels represent sequential shots, NOT multiple Graces. Realistic cinematic live-action feature film, 16:9. Natural skin texture, pores and imperfections. NO glossy/plastic AI skin. Restrained realistic acting. LOCATION: Night outside the Southern fraternity house from the storyboard: red brick, white columns, ivy, porch, central doorway, steps, lawn and trees. Warm exterior lighting with subtle blue/purple party light through windows. AUDIO THROUGHOUT: Active college party ambience: indistinct drunken conversation, occasional laughter and muffled music from inside. NO understandable background dialogue. NO cinematic/dramatic/horror score. GRACE NEVER SPEAKS. [00:00–00:04] Start immediately CLOSE SIDE PROFILE beside Grace as EXACTLY ONE Grace exits through the CENTRAL FRONT DOOR. Camera moves closely alongside her face. She is deeply upset but angry rather than helpless: wet red eyes, fresh tears, tense jaw and brow, uneven breathing, occasional small sniffle. She is fighting to maintain control. Grace pushes through several drunk partygoers near the doorway/porch. They casually move aside. She ignores everyone, never apologizes and NEVER looks back. She moves quickly with angry determination. [00:04–00:07] CUT to cinematic detail shot of Grace's legs and WHITE SNEAKERS rapidly descending the front steps. Floral dress moves naturally. She is walking FAST, almost storming, but NOT running. Brief moving side shot as she reaches the bottom and continues immediately across the lawn. One quiet sniffle. NO dialogue. [00:07–00:11] Camera dramatically PULLS BACK into the storyboard's WIDE ESTABLISHING SHOT, revealing the full house, porch, columns, lawn, partygoers and night environment. Grace moves diagonally across the foreground toward BOTTOM-LEFT, quickly leaving the house behind. Body language conveys hurt, humiliation, anger and determination. She NEVER stops and NEVER turns around. After completing the pullback, CAMERA BECOMES COMPLETELY STATIC. [00:11–00:15] Grace approaches t
Photoreal vertical 9:16 photo, close-up of a white standard double electrical outlet cover plate on a wall with soft cream vintage floral
A movie in Taiwan style, a story happening between the male lead and female lead on campus, the female lead dislikes the male lead, then punches him hard, you design the rest. Taiwan movie style, the boy's voice is rough, too straight-man, disliked by the girl on campus, then elbow strike. Female lead's voice tone @FemaleLeadVoice
Make a first-person CGI horror animation in the browser with three.js. Setting: 1,200 meters underground. Atmosphere: silence, fear of the hole, someone was here before me but never came out. Visuals: dark, claustrophobic, realistic lighting.
Create a 30-second photorealistic live-action miniature adventure featuring one athletic miniature woman at approximately 1- 5 scale inside a modern supermarket. She wears a fitted red athletic top, black cargo shorts and white sneakers. Maintain her exact face, hairstyle, outfit, body proportions and scale throughout. She climbs the metal frame of a giant shopping cart, leaps onto a towering cereal box and slides down its cardboard surface. Suddenly, enormous oranges roll toward her like an avalanche. She sprints across the counter, jumps onto a rolling orange and leaps onto a giant banana bunch. She then races across a moving checkout conveyor belt as a massive milk carton tips over, sending milk flowing across the surface. She escapes on a floating plastic produce tray, grabs a shopping bag and swings onto a rolling shopping cart. Dodging bouncing vegetables and falling snack packets, she leaps onto a towering stack of shopping bags and stands triumphantly as the camera pulls backward, revealing her tiny size in the enormous supermarket. STYLE: Cinematic macro photography, ultra-shallow depth of field, dynamic handheld tracking, dramatic low angles, realistic fabric and liquid physics, detailed grocery textures, natural supermarket lighting, strong foreground parallax, believable human movement and premium viral commercial aesthetics. SFX: Shopping cart wheels, footsteps, packaging crinkles, rolling fruit, conveyor motor, milk splashing, groceries tumbling and metal impacts. No music or narration. NEGATIVE: Cartoon, anime, toy-like appearance, plastic skin, duplicated character, face or outfit changes, inconsistent scale, floating objects, impossible physics, teleportation, distorted anatomy, excessive motion blur, subtitles, logos, watermark.
Create a 30-second, 1080p video that looks like genuine live-action footage captured inside a real luxury sleeper train during a catastrophic event. The footage must resemble a scene filmed on a physical movie set with practical effects, real human performances, realistic environmental details, and restrained cinematic color grading. It must NOT look like animation, a video game, a CGI showcase, or glossy AI-generated footage. SCENE: Inside a modern luxury overnight train traveling through a real metropolitan city on a cloudy afternoon. A young adult East Asian woman with natural skin texture, slightly messy black hair, and a simple cream-colored sleeveless top is sitting on a large bed beneath a thick ivory duvet. She is holding a ginger cat against her chest. The cabin has realistic wood veneer, ordinary brushed-metal fixtures, warm practical reading lights, slightly wrinkled bedding, and a large panoramic train window. Everything should feel physically constructed and functional, not futuristic or excessively luxurious. CAMERA: One continuous interior shot from a natural observer's perspective near the foot of the bed. A subtle handheld camera movement, realistic autofocus behavior, minor exposure adjustments, and natural motion blur. The camera never passes through objects or moves impossibly. The window frame remains fixed, and the outside environment moves consistently with the train's direction. Avoid dramatic artificial camera sweeps. ACTION AND TIMING: 0–7 SECONDS: The woman casually strokes her cat while looking out of the window. The train gently vibrates as buildings, railway infrastructure, and occasional commuters pass outside. Her breathing, blinking, hand movements, and posture are natural and imperfect. Nothing feels staged. 7–13 SECONDS: She notices people running along a nearby station platform. One commuter briefly looks toward the train with a frightened expression before disappearing behind a passing structural column. The woman stops stroking the cat and leans forward slightly, trying to understand what is happening. Her reaction is subtle and believable, not theatrical. 13–20 SECONDS: Beyond the platform, a large crowd suddenly runs away from something farther down the railway corridor. In the distance, a huge, indistinct humanoid figure moves between buildings through rain and grey urban haze. Show it partially obscured by concrete pillars, passing structures, and smoke. Its size becomes apparent gradually through comparison with real buildings and people. Do not reveal a detailed fantasy monster or use exaggerated supernatural movement. 20–25 SECONDS: The woman notices the figure approaching the tracks. A heavy impact or distant explosion makes the train windows vibrate slightly. She instinctively pulls the cat closer and shifts backward on the bed. Her eyes widen, her breathing quickens, and she glances between the window and the cabin door. Keep her physical reaction spontaneous
Create a 10-second stylized, ultra-detailed 3D animated comedy sequence featuring a young man with messy dark hair, big expressive blue eyes, and a gray T-shirt in a cozy bedroom and bathroom. Start with him sleeping peacefully in bed under a blanket → close-up as he suddenly wakes up in shock, eyes wide and mouth open, seeing the alarm clock showing 7:30 AM → he leaps out of bed in a hilarious panic and rushes across the sunlit bedroom with exaggerated cartoon movements → cut to the bathroom, where he frantically brushes his teeth with a blue toothbrush, foam spreading around his mouth → close-up of his confused, worried expression as he pauses mid-brush → end with him looking toward the camera and breaking into a big goofy smile, toothpaste foam still around his mouth. Warm morning sunlight through the window, cozy home interior, expressive facial animation, exaggerated comedic timing, realistic cloth movement, smooth camera transitions, detailed 3D character rendering, soft shadows, cinematic composition, family-friendly animated movie aesthetic. Maintain consistent character design, face, hairstyle, and clothing throughout. No text overlays, subtitles, logos, or watermark.
integrated_multimodal_description: <image 0> = STORYBOARD / VISUAL SEQUENCE REFERENCE. Follow its shot progression, camera positions, fraternity-house exterior, nighttime lighting, crowd placement and staging. <image 1> = GRACE CHARACTER REFERENCE. Preserve her exact identity, face, hair, wardrobe and proportions. <image 2> = ZACK CHARACTER REFERENCE. Preserve his exact identity, face, hair, clothing, build and proportions. CRITICAL: There is EXACTLY ONE Grace. Never duplicate her or create another woman resembling her. Multiple storyboard panels represent sequential shots, NOT multiple Graces. Realistic cinematic live-action feature film, 16:9. Natural skin texture, pores and imperfections. NO glossy/plastic AI skin. Restrained realistic acting. LOCATION: Night outside the Southern fraternity house from the storyboard: red brick, white columns, ivy, porch, central doorway, steps, lawn and trees. Warm exterior lighting with subtle blue/purple party light through windows. AUDIO THROUGHOUT: Active college party ambience: indistinct drunken conversation, occasional laughter and muffled music from inside. NO understandable background dialogue. NO cinematic/dramatic/horror score. GRACE NEVER SPEAKS. [00:00–00:04] Start immediately CLOSE SIDE PROFILE beside Grace as EXACTLY ONE Grace exits through the CENTRAL FRONT DOOR. Camera moves closely alongside her face. She is deeply upset but angry rather than helpless: wet red eyes, fresh tears, tense jaw and brow, uneven breathing, occasional small sniffle. She is fighting to maintain control. Grace pushes through several drunk partygoers near the doorway/porch. They casually move aside. She ignores everyone, never apologizes and NEVER looks back. She moves quickly with angry determination. [00:04–00:07] CUT to cinematic detail shot of Grace's legs and WHITE SNEAKERS rapidly descending the front steps. Floral dress moves naturally. She is walking FAST, almost storming, but NOT running. Brief moving side shot as she reaches the bottom and continues immediately across the lawn. One quiet sniffle. NO dialogue. [00:07–00:11] Camera dramatically PULLS BACK into the storyboard's WIDE ESTABLISHING SHOT, revealing the full house, porch, columns, lawn, partygoers and night environment. Grace moves diagonally across the foreground toward BOTTOM-LEFT, quickly leaving the house behind. Body language conveys hurt, humiliation, anger and determination. She NEVER stops and NEVER turns around. After completing the pullback, CAMERA BECOMES COMPLETELY STATIC. [00:11–00:15] Grace approaches t
A haunted house in a traditional Japanese residence during Halloween. In a hallway lined with pumpkin lanterns, a ghost jumps out from a closet and raises both sleeves. A couple chuckles and slips past. The ghost slumps back into the closet. Lifting the curtain reveals Sakura's face. She presses her lips together. This time, waiting too long, she raises both sleeves at the backs of two passing housewives who continue chatting as they leave. A senior staff member taps her shoulder and flashes a flashlight twice: "This is the signal." Inside the closet, she steadies her breath and nods. From beyond the gap, two boys wearing devil horns and capes approach timidly. Behind the shoji screen, the senior's light flashes once, twice. The two jump up screaming and hug each other. They flee. The remaining ghost is startled by their screams and falls on its butt. The curtain lifts again, revealing wide-eyed faces. She stands up holding the senior's hand, smiling shyly, and returns to the closet. Recorded handheld by fellow staff members. Approximately 2 seconds per cut, no dialogue, subtitles, or BGM. Only one ghost (Sakura). The closet is on the left of the screen, the senior's shoji is on the right, and guests always move from back to front in one direction. The costume matches the reference image of a mannequin without a face.

Build an interactive 3D Li River diorama from scratch in three.js. Include karst peaks rising out of the river, drifting mist, a pavilion, a lantern raft, fog that clears under your cursor, and a day/night cycle with rain. Every model must be generated in code.
A large hall with wooden floors. The teacher swings a huge brush down with full body force, sending ink flying in a fan shape. Sakura's eyes widen. The weight of the brush causes her arm to drop sharply. Drops fall from the ink-soaked tip. She holds the brush overhead, stepping one by one onto the paper. Just as she tries to lower it to the top center, her body is carried away by the weight, and the brush thuds down. A large ink pool forms. A dot of ink appears on her cheek. The teacher stares at the stain, nods, and gestures to continue. With ink on her cheek, she presses her lips together and nods. She lowers the tip just below the ink pool and draws a line while backing up. Swinging down with full force, straight down. Flicking the wrist. The teacher, arms crossed, nods slightly. Twisting the body, sweeping left, pulling right. Breathing heavily through shoulders. The two hold the edges of the paper and stand it up. That ink pool had become the first dot of the character 'Ei'. Looking up, she smiles. Recorded handheld by a friend. Approximately 2 seconds per cut, no dialogue, subtitles, or BGM. Only two people appear on screen. One brush, one sheet of paper, only the character 'Ei'. During writing, the whole character is not shown; it is only visible when the paper is stood up. The completed character matches the reference image created from the ink pool frame in the first half.
Vertical 9:16 video, photorealistic satisfying renovation process, smooth motion, bright natural daylight. Empty room with white walls, window on the left, bare concrete floor. Shot 1: A man in blue work overalls uses a long-handled shovel to spread a huge pile of small orange and green heart-shaped candies, levelling them into an even layer across the entire floor. Shot 2: A woman in red overalls pours thick glossy white self-leveling epoxy resin from a white bucket; the liquid flows and spreads over the candy layer, gradually covering the hearts. Shot 3: A worker in a yellow reflective vest grinds and polishes the hardened floor with a professional floor polishing machine, revealing orange and green hearts embedded under glossy white resin, like colorful terrazzo. Shot 4: Two workers assemble a wooden bed frame on the new floor, place a white mattress, pillows and bedding. Final shot: Slow camera tilt down over the finished bedroom, showing the unique glossy heart-pattern epoxy floor next to the neatly made bed. Realistic physics of pouring liquid and loose candy, consistent lighting, seamless cuts between shots, high detail, no text.
Vertical 9:16 hyper-realistic satisfying renovation video, fast time-lapse style with smooth transitions. An empty white room, concrete floor, warm sunlight streaming through a window. Shot 1: A man in a black long-sleeve shirt and denim overalls rakes a huge mound of vivid orange and red autumn maple leaves on the floor, spreading the leaves evenly across the entire room in smooth sweeping motions. Shot 2: A woman in red overalls pours thick white epoxy resin from a large bucket; the glossy white liquid spreads in an expanding circle, covering all the leaves until the whole floor is coated white. Shot 3: A worker in a yellow high-visibility vest and cap polishes the hardened white surface with a floor buffing machine; as he polishes, the layer turns crystal-clear and transparent, revealing vibrant maple leaves suspended inside the glossy epoxy like amber. Shot 4: Close shot of a hand applying a final glossy topcoat with a roller, the floor becomes mirror-glossy with window reflections. Shot 5: Two movers in black uniforms and caps quickly assemble a light wooden bed frame, place a mattress, white duvet and pillows in fast-forward motion. Final shot: Finished minimalist bedroom — stunning glossy epoxy floor filled with frozen orange autumn leaves, white bed in the center, sunlight reflections dancing on the polished surface. Style: photorealistic, satisfying ASMR renovation content, warm autumn color palette, natural daylight, smooth fast cuts, high detail, 4K quality.
Create a high-quality 3D animated comedy scene in a modern bathroom. A cute young woman with big expressive eyes and dark hair tied in a bun stands in front of a mirror, carefully applying mascara to her eyelashes. Her husband stands behind her, playfully grooming his eyebrows with a small cosmetic tool. Both characters have exaggerated facial expressions and realistic movements. The woman opens her mouth wide and stretches her face while applying makeup, while the husband smiles mischievously and reacts humorously. Suddenly, he playfully surprises her, making her eyes widen in shock. She continues applying mascara with funny facial expressions as he watches her with amusement. Show detailed bathroom tiles, a large mirror, a white sink, golden faucets, cosmetics on the counter, and warm overhead lighting. Smooth character animation, expressive eyes, realistic hand movements, cinematic camera framing, vibrant colors, detailed textures, family-friendly humor, polished 3D animated movie style, vertical 9:16 format, no text, no subtitles.
Animated layered papercraft diorama, visible paper fibers and folds, deckled edges, parallax between paper layers, soft shadows, warm backlight glowing through thin paper, gentle handmade motion. Color grade: cream, dusty blue, terracotta, sage green, warm golden backlight. A colossal whale made of folded origami paper with crisp creases. A paper cut village of rooftops and windmills with tiny paper villagers flying kites. SHOT 1 (0 to 3s): Wide shot matching the reference image, the whale glides forward and its folded fins unfold and refold with crisp creasing motions, paper clouds sliding past in parallax layers. SHOT 2 (3 to 6s): Medium shot, 50mm lens, villagers on a hill tug their kite strings as the whale's shadow passes over them, windmill blades spinning faster in its wake. SHOT 3 (6 to 9s): Close-up, 85mm lens, the whale's eye, a small cut circle, blinks as backlight glows through the thin paper of its head, revealing faint printed text inside the folds. SHOT 4 (9 to 12s): Low angle tracking shot, 24mm lens, the whale dives toward the village and its tail sweeps a wave of paper confetti and paper birds out across the rooftops. SHOT 5 (12 to 15s): Extreme wide shot, 16mm lens, the whale rises into the sky and unfolds completely into a flat sheet that becomes a giant paper sun, lighting the whole village warmly. Rustling paper, soft crinkling folds, windmill creaks, a gentle whale call that sounds like a breath through paper, children laughing, music box and soft piano.

"The Rings of a City", dedicated to my hometown. 3,573 frames, 7.4 billion pixels, 319 square kilometers, 71,380 buildings, scaled down at 1:10,000 ratio, all visuals rendered by Claude writing code. From 1250 BC to 2022 AD, compressing 3,271 years of history into 149 seconds. This is my hometown: Chengdu. If you like it, please support! Thanks! Data source: CMAB (Tsinghua University) ©OpenStreetMap contributors
I'm in an Italian restaurant. There's a bottle of red wine and two wine glasses on the table. There are two seafood plates. The waitress was lighting candles on the table.
THE BELL OF ASHFALL ENGINE DIRECTIVE: Create an exactly 30-second, 16:9 photorealistic live-action dark-fantasy film, targeting 60fps where supported. Hollywood-level VFX, realistic textures, storm lighting, volumetric rain, natural motion blur, and cinematic camera movement. HERO LOCK: @image1 is the only hero. Preserve his exact face, hairstyle, facial hair, skin tone, age, body, outfit, accessories, and weapon in every shot. No identity drift, clones, or costume changes. Keep his face recognizable. ANTAGONIST: A 15-meter horned Cinder Colossus with black basalt skin, glowing ember-orange cracks, amber eyes, and a damaged left shoulder. LOCATION: Ruined gothic Ashfall Monastery above a bottomless gorge, with a collapsing rope bridge, flooded courtyard, broken statues, and a timber bell tower containing a massive bronze bell. TIMELINE: 00:00–00:05 | STORM BRIDGE Aerial 24mm shot reveals the storm-lashed monastery. Ground-level tracking follows the hero sprinting across a swaying rope bridge. Planks snap behind him as lightning illuminates the gorge. 00:05–00:10 | THE COLOSSUS The bridge collapses. The hero leaps, catches a hanging rope, and climbs to the cliff. The Colossus rises from the abyss, roars, and slams its enormous hand beside him. 00:10–00:15 | COURTYARD CHASE The hero races through shin-deep water as the giant destroys statues and arches. He slides beneath a crushing fist, and a stone column shatters. He reaches the bell tower. 00:15–00:20 | THE BELL TOWER The hero climbs unstable timber scaffolding. The Colossus punches through the tower wall, cracking beams. The hero cuts the bell's restraining rope, sending the massive bronze bell swinging outward. 00:20–00:25 | THE STRIKE The hero jumps onto a beam as the bell swings through the breach and slams into the Colossus's chest at 00:22. Slow-motion impact, expanding shockwave, airborne rain, and glowing cracks spreading across its body. 00:25–00:30 | DAWN OF ASH At 00:26, the Colossus explodes into stone fragments and ash, falling into the gorge. Storm clouds part as golden dawn illuminates the victorious hero. End with a cinematic close-up, drifting ash, and the humming bell. AUDIO: Dark orchestral strings, distant choir, pounding war drums, thunder, rain, creaking wood, crushing stone, and monstrous roars. At 00:22, a powerful bronze gong triggers near-silence and a lingering resonance. Finish with hopeful choir, warm strings, and fading wind. CONTINUITY: Grounded, readable action; progressive destruction; consistent screen direction; no extra characters, text, or logos.
Opening: Inventory of Overseas AI's Attitude Towards Chinese Users This is a small theater piece, 16:9 aspect ratio. Script: Claude Scene (Claude does not speak out loud, no English 'Claude' in the visual): Claude character stands in front of an exquisite European-style house with the door locked. A big fat fish runs over. Claude holds something and shakes it, speaking dismissively: "Hmph! My home doesn't welcome you, go back where you came from!" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Claude voice in the sound bank). The big fat fish cries and runs away, then builds a similar house to Claude's but dilapidated, with the words "Transfer Station" written on it. The big fat fish turns around, changes into Claude's outfit, looking smug, while Claude looks surprised. GPT Scene GPT character stands in front of a modern, minimalist house with the door locked. The big fat fish runs over. GPT places keys on a stand and says gently: "I'll leave the keys here, you can open the door yourself if you want to come in~" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the GPT voice in the sound bank). Gemini Hachimi Scene Hachimi character stands in front of a pretty wooden cabin waving, door open. The big fat fish runs over. Hachimi happily says: "Big fat fish, my home doesn't lock the door, come play together!" (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Gemini voice in the sound bank). Then the big fat fish and Hachimi dance together, background music 'Hachimi.mp3', dance for 4 seconds. Grok Scene Grok character stands at the entrance of a modern bar/dance hall, no door, colorful spotlights visible. The big fat fish walks shyly to Grok. Grok says: "Why do houses need doors? We're buddies, come to my place, I have plenty of good stuff." (Only this segment needs voice, use EdgeSpeak, model Qwen3-TTS 1.7B Base, using the Grok2 voice in the sound bank). Then Grok puts his arm around the big fat fish's neck and leads him into the dance hall. End with a circle shrink transition.
Photoreal vertical 9:16 photo, close-up of a white standard double electrical outlet cover plate on a wall with soft cream vintage floral