Follow + Retweet and Comment "Prompt" to get Master promt used for this workflow 💯
Tools used in this video (all free):
Claude Code (Free Plan): zapiwala.ai/s/y1ME5
ElevenLabs (Free AI Voice Generator): zapiwala.ai/s/ipR4S
FoziScribe AI (Free Voiceover to Timestamp Script): zapiwala.ai/s/jLyYh
Google Flow (Free AI Image & Video Generator 2026): zapiwala.ai/s/ieVLk
Bulk Image Generator (ZAPI FLOW 100% Free Forever Chrome Extension): zapiwala.ai/s/ubeoz
Cukong Crypto retweeted
ESTO ES JODIDAMENTE LOCO
Este tipo explicó cómo creó 3 sitios web que le generan $130,000 al mes
En 25 minutos, muestra cómo lo hizo, paso a paso. Totalmente gratis.
Idea → Web → Monetizar
Guarda esto, me lo agradecerás después 🔖
Cukong Crypto retweeted
🚨ESTO ES UNA LOCURA
Un desarrollador chino acaba de lanzar una herramienta GRATUITA capaz de crear vídeos automáticamente para TikTok, Reels y YouTube Shorts.
Se llama MoneyPrinterTurbo.
Ya ha superado las 100.000 estrellas en GitHub.
Y ya hay gente ganando +4.000$/mes con canales faceless a partir de este sistema.
Así funciona:
1. Introduces un tema o una palabra clave
2. La IA genera el guion, la narración y los subtítulos
3. Busca el material visual, edita el vídeo y te entrega el resultado final
Todo ocurre en un único flujo de trabajo.
Guárdate esta herramienta.
Estoy seguro de que puedes sacarle muchísimo partido.
Enlace abajo👇
Cukong Crypto retweeted
Monday calling ☎️
Seedance 2.0 and GPT Image 2
Prompt :
Use the uploaded reference image as the exact character reference. Preserve her facial identity, hairstyle, eye color, makeup, skin tone, body proportions, and facial consistency throughout every shot. Maintain the same character consistently. Create a 15-second ultra-realistic GRWM office routine with 15 fast-paced cinematic shots (≈1 second each), smooth match cuts, whip pans, object transitions, premium lifestyle cinematography, and realistic motion.
1. Alarm rings as she opens her eyes.
2. She sits up, stretches, and smiles.
3. Quick skincare splash in the bathroom.
4. Applies light makeup in front of the mirror.
5. Brushes and styles her hair.
6. Changes into a chic office outfit.
7. Puts on heels, watch, and earrings.
8. Grabs her handbag, laptop, and coffee tumbler.
9. Locks the apartment door behind her.
10. Walks confidently through the apartment lobby.
11. Hails a cab on a busy city street.
12. Rides through downtown with cinematic city reflections.
13. Steps out in front of a modern glass office building.
14 Scans her access card and enters the office.
15 Sits at her desk, opens her laptop, smiles at the camera, and begins working.
Ultra-realistic commercial quality, smooth handheld and gimbal camera movement, natural expressions, realistic body movement, premium office interiors, luxury apartment, cinematic color grading, shallow depth of field, realistic reflections, 4K HDR, 16 9, no text, subtitles, logos, or watermarks.
Cukong Crypto retweeted
I’m genuinely surprised how many people still struggle with AI image generation. It’s not just about writing a prompt, the right workflow can completely change the results.
Here’s the complete AI image generation guide
Cukong Crypto retweeted
【AIセクシーダンス決定版プロンプト】
セクシーで自然な体の動きができるSeedance2.0の動画プロンプトの現時点での最適解がこちら!
だいぶ腰の動かし方が自然になった気がする!!
K-POPでコンサートのソロダンスパートで踊るセクシーダンスのイメージです😎
<プロンプト>
The exact same girl from the reference image @ 画像1— 100% identical face, hairstyle, outfit, body proportions, no changes whatsoever, clean idol skin with subtle glow.
She performs intense K-POP girl-crush dance break: aggressive hip shakes, deep body rolls, fluid body waves, thigh isolations, hair whipping wildly, deadly sexy facial expressions.
Multi-cut music video setup: starts with a smooth 360° tracking shot around her body, dynamic dolly zoom, then cuts into a fetishistic montage of extreme close-ups showing her hip thrusts, waist isolations, and sharp choreography, ending with a dramatic low angle push in to her face.
Cinematic color grading: warm skin tones with deep teal shadows, subtle film grain, high-end music video style.
fit the attached character image.
4K, photorealistic, ultra-sharp skin and fabric details, mixture of slow-motion and real-time, high-end fashion magazine advertisement style, high-end MV style, 24fps.
🤖 Made with AI
Cukong Crypto retweeted
Seedance 2.0 on @pixverse
Short prompt version
================
Prompt:
Night retro old-town street, cinematic CG, high-fashion girl group MV mood, shallow depth of field, cool-warm color grade.
0-2s: Medium shot, slow arc left. [image2] stands under a streetlamp, lifts her eyes to camera. Focus locked on [image2].
2-5s: Camera keeps arcing left, focus pulls forward. Blurred figure [image2] in the foreground resolves into a low-angle close-up, turning to face camera. [image2] blurs into bokeh.
5-9s: Camera pulls back, panning left. [image3] recedes to mid-ground. [image4] walks in from the left, filling the foreground. Focus shifts from [image3] to [image4].
9-12s: Camera arcs around. [image4]falls back. [image1] walks in from the right into a foreground close-up. Focus moves to [image1].
12-15s: Wide shot, camera pulls back and rises, then settles. Four members line up in a power stance — [image2] center, [image3] and [image4] on the sides, [image1] closing. Stone arches for depth, streetlamps forming a light trail, wet reflective ground.
================
#PixVerseCPP
🤝 Paid partnership
Cukong Crypto retweeted
I'm amazed most people still spend hours learning AI video creation through trial and error.
Here's the complete AI video generation guide that helps you create cinematic videos like a pro.
---
1. Learn what AI video generation is and how it transforms text, images, audio, and videos into new content.
---
2. Discover the best AI video models, including Seedance 2.0, Veo 3.1, Kling 3.0, Runway Gen-4, Hailuo, Wan 2.6+, Pika, and Luma Ray.
---
3. Understand every major type of AI video generation, from text-to-video to cinematic video creation.
---
4. Follow a proven prompt formula covering subject, action, location, lighting, camera movement, style, motion, audio, and quality.
---
5. Use detailed prompts instead of vague instructions for dramatically better results.
---
6. Learn popular camera movements like FPV, crane, orbit, tracking, zoom, and dolly shots.
---
7. Choose the right camera lens to create the exact cinematic look you want.
---
8. Experiment with visual styles like cinematic, photorealistic, anime, Pixar, cyberpunk, and fantasy.
---
9. Improve realism with lighting styles such as golden hour, blue hour, studio light, fog, and moonlight.
---
10. Follow the complete AI video workflow from idea to publishing.
---
11. Create videos for YouTube, Instagram Reels, TikTok, marketing, storytelling, education, and product demos.
---
12. Edit AI videos by removing objects, replacing backgrounds, extending clips, and upscaling quality.
---
13. Add AI voiceovers, captions, music, motion tracking, and sound effects with ease.
---
14. Use the correct aspect ratio for every platform before exporting.
---
15. Follow an export checklist to ensure smooth motion, synced audio, captions, and final quality.
---
16. Avoid common mistakes like short prompts, conflicting instructions, and missing lighting details.
---
17. Keep one clear action per shot to generate more consistent videos.
---
18. Test multiple AI models because each produces different creative results.
---
19. Save your best prompts and refine them after every generation.
---
20. The infographic includes even more workflows, examples, best practices, and optimization tips.
---
To learn more, check the infographic.
---
Cukong Crypto retweeted
Seedance 2.0 on pollo ai
Try it with prompt :
15 seconds | 16:9 | live-action military drama
Fictional dramatization. Documentary-style visual realism.
SETTING
A fictional frontline environment during an Iran–Israel military confrontation.
[desert / damaged urban outskirts / mountain road / military checkpoint].
Time: [dawn / midday / dusk / night].
Dry air, dust, distant smoke and realistic background activity.
FORCES
Two clearly distinguishable military groups representing opposing Iranian and Israeli forces.
Maintain consistent uniforms, equipment, vehicles and character identities throughout the sequence.
Human behaviour remains grounded and restrained.
STORY
00–04s — ESTABLISHING
Wide observational shot establishing both sides separated across the environment.
Personnel move into positions while vehicles and environmental activity remain visible in the distance.
Camera slowly advances from a believable documentary position.
Natural wind moves clothing, dust and vegetation.
00–08s — TENSION
Cut to a closer handheld perspective following personnel moving through cover.
They communicate through gestures and short commands.
Camera operator follows behind with natural footsteps, small framing corrections and realistic motion blur.
Dust reacts naturally to movement.
08–12s — CLASH
The confrontation begins.
Show confusion, rapid movement, personnel taking cover and environmental reactions from distant impacts.
Keep geography readable so both sides remain spatially consistent.
Camera reacts naturally to sudden events rather than anticipating them.
Avoid graphic injuries or gore.
12–15s — AFTERMATH
Camera settles behind cover.
Dust and smoke drift through the environment while distant movement continues.
One person cautiously looks toward the opposing position.
End on unresolved tension rather than a stylized victory pose.
CAMERA
Grounded conflict-documentary cinematography.
Mix:
wide environmental coverage
medium handheld following shots
brief close reaction shots
Natural handheld instability.
Realistic operator movement.
Plausible camera positions.
No impossible drone transitions.
No excessive slow motion.
LIGHTING
Lighting originates from the actual environment.
Natural exposure changes when moving between sunlight, shadow, smoke and interiors.
Realistic highlight roll-off.
Deep but detailed shadows.
No artificial rim lighting around characters.
PHYSICS
Real gravity and momentum.
Feet maintain contact with terrain.
Equipment has believable weight.
Dust reacts to footsteps and vehicle movement.
Clothing responds to wind and body movement.
Smoke follows environmental airflow.
Characters physically interact with walls, vehicles and terrain.
HUMAN PERFORMANCE
Restrained, believable behaviour under extreme stress.
Heavy breathing.
Quick visual checks of surroundings.
Small posture adjustments.
Natural hesitation and urgency.
No heroic posing or exaggerated action-movie gestures.
AUDIO
Dry wind and distant environmental ambience.
Boots against gravel and concrete.
Equipment and clothing movement.
Vehicle engines at realistic distance.
Short shouted commands.
Distant impacts with appropriate acoustic delay and environmental reflections.
No oversized Hollywood trailer effects.
VISUAL CHARACTER
Photorealistic live-action footage.
Observational war-documentary character.
Natural skin texture.
Realistic fabric and material response.
Subtle sensor noise.
Natural motion blur.
Imperfect handheld framing.
Avoid:
video-game aesthetics
CGI appearance
excessive HDR
oversaturated fire
impossible explosions
floating debris
perfectly choreographed movement
beauty-filter faces
graphic gore
CONTINUITY
Maintain character identity, uniforms, equipment, terrain and lighting between shots.
Preserve geographic orientation and screen direction.
Every action begins from the physical state established by the previous shot.
🤖 Made with AI
趣味AI视频 -制作流程:
提示词生成:把提示词和你的宠物照片 发给 Codex,让它基于图片 修改为一段新的提示词,剧情和构图不变
视频生成:豆包 选择 Seedance 2.0 mini ,输入Codex新生成的提示词和你家的宠物照片,时长选择5秒
生成视频:【风格】超写实冰淇淋店魔术恶作剧短视频(Photoreal Magical-Realism Gelato Shop Video)。冰淇淋柜中躺着一只真实银灰英国短毛猫,严格参考上传图片的圆脸、银灰虎斑、金黄色眼睛和厚脸颊;模拟用冰淇淋挖勺从它身上挖出一撮松散猫毛,再放进华夫蛋筒,毛团在蛋筒内瞬间变成迷你写实小猫。全程是无伤害魔术效果,不接触皮肤、不拔毛、不造成疼痛。竖屏9:16,5秒,真实手机近景,无字幕、无水印。
【场景】明亮温暖的手工冰淇淋店玻璃展示柜。左侧不锈钢盘是粉色草莓冰淇淋,右侧盘是浅黄色香草冰淇淋;中间是一只室温展示道具盘,底部铺米色软垫,银灰英短舒服地仰躺其中,像一大球银灰冰淇淋。画面有玻璃反光、金属边框与展示柜冷光。
提示词👇👇👇
【角色】 大猫:真实成年银灰英短,圆脸、金黄色圆眼、银灰渐层虎斑、浅色嘴周;仰躺时四只爪子自然蜷起,神情淡定放松。 迷你小猫:约成人拇指大小,是大猫的写实缩小版,同样有银灰虎斑、圆脸、金黄色眼睛和浅色嘴周。
【道具】 蓝色一次性手套、不锈钢圆形冰淇淋挖勺、一个金黄色华夫蛋筒。蛋筒口朝上,边缘有酥脆锯齿纹,内部为空。
【声音】展示柜轻微冷气声、挖勺轻柔“沙沙”声、大猫舒服呼噜声;瞬间变身时一声短促干净的“啵”,随后迷你小猫奶声“喵”。
[00:00-00:01.5] 镜头1:像挖冰淇淋一样取毛团 镜头俯拍展示柜中央。银灰英短仰躺在软垫展示盘中,腹部上方有一小撮自然浮起的松散银灰猫毛。蓝色手套拿圆形冰淇淋挖勺轻轻旋转,将浮毛卷成一颗圆滚滚的银灰毛团;大猫身体和毛发仍完整平滑,没有秃斑,舒服地眨眼呼噜。
[00:01.5-00:03.5] 镜头2:毛团放进华夫蛋筒 另一只蓝色手套拿着空的金黄色华夫蛋筒,蛋筒口朝上。持挖勺的手把银灰毛团从挖勺中轻轻倒进蛋筒深处,毛团刚好填满蛋筒顶部。此时蛋筒里仍然只是一颗蓬松的银灰毛球,绝不提前出现小猫。
[00:03.5-00:05] 镜头3:瞬间变成小猫 毛团落入蛋筒后的0.2秒内瞬间完成变形:伴随一声短促“啵”,蓬松毛球立刻变成一只拇指大小、完全写实的迷你银灰英短,没有缓慢生长过程。小猫立刻从蛋筒口探出圆脸,两只小耳朵竖起,金黄色大眼睁开,一只小爪搭在蛋筒边缘,奶声“喵”一下。手托着蛋筒举向镜头,背景中的大猫仍安稳躺在展示盘里,画面可爱收尾。
【负面约束】 真实冰淇淋接触猫、低温冻伤、拔毛、剃毛、刮皮、伤口、血液、疼痛、猫挣扎、秃斑、锋利刀具、缓慢变形、变形过程拉丝、猫被硬塞进蛋筒、窒息、恐怖感、卡通感、文字、水印。,9:16
🤖 Made with AI
Cukong Crypto retweeted
What if fantasy epics could be brought to life with AI?
"The Celestial Slayer" is an AI-generated cinematic trailer following the journey of a celestial queen who transforms into a divine warrior, blending breathtaking world-building, magical visual effects, and blockbuster-inspired action into a fantasy adventure.
Created entirely using next-generation AI creative workflows, this project demonstrates how creators can produce cinematic trailers, concept films, and immersive visual stories without the traditional limitations of large-scale film production.
The future of fantasy filmmaking is being reimagined,one frame at a time.
What do you think? Would you watch a full-length AI-generated fantasy film like this?
Story board and prompt and in the comments
Tools used:
• ChatGPT — Character
• Seedance 2.0 — Video generation
• CapCut — Editing
Cukong Crypto retweeted
I made this 15-second clip using Seedance 2.0, with GPT Image 2 on @higgsfield_ai
Prompt seedance 2.0 :
Use the supplied image1 as the exact first frame of the video.
Maintain perfect visual continuity with the reference image. Preserve the exact young woman, the exact robot soldier, the industrial facility, wardrobe, proportions, facial identity, robot construction, lighting, color palette, camera perspective, and environment. No redesigns or substitutions.
Style:
Authentic late-1980s live-action sci-fi comedy filmed on 35mm Kodak motion picture film. Practical sets only. Full-size practical mechanical robot suit with visible hydraulic joints and realistic mechanical weight. Warm tungsten industrial lighting, subtle atmospheric steam, soft halation, natural film grain, slight gate weave, vintage telecine color, practical smoke, cinematic depth of field.
Scene:
The woman and the robot casually walk side by side through the industrial facility. Their pace is relaxed and natural. The robot moves with heavy hydraulic weight, subtle servo corrections, piston movement, and believable mechanical inertia. The woman walks confidently while casually talking, occasionally looking up at the robot, smiling, and using expressive hand gestures. Their interaction feels friendly and comedic rather than exaggerated.
Camera:
Continuous cinematic tracking shot.
Medium two-shot transitioning into a smooth side tracking shot.
Steady dolly movement only.
No handheld shake.
No whip pans.
No sudden zooms.
No jump cuts.
Maintain eye-level framing throughout.
Background:
Industrial pipes, catwalks, warm practical lights, drifting steam, distant workers performing routine tasks without interacting with the main characters. Background remains secondary and never distracts from the conversation.
Timeline
0:00–0:04
Medium two-shot.
The woman and robot begin walking together.
The woman glances at the robot with a playful smile and says in a cheerful 1980s California accent:
WOMAN:
"I'm seriously jealous of all the FLUX 3 cool kids."
Natural conversational lip sync.
0:04–0:10
Smooth side-tracking shot as they continue walking past pipes and drifting steam.
The woman becomes more animated, making larger but natural arm gestures while speaking:
WOMAN:
"Their UNHINGED FLUX 3 videos have been everywhere all weekend!"
Maintain realistic walking speed and synchronized body movement.
0:10–0:13
Natural push-in to a close-up of the robot.
The robot slowly turns its head toward her with believable hydraulic motion and a soft mechanical servo whir. Its eye panel briefly flickers before speaking in a dry synthesized voice:
ROBOT:
"I know, right?"
The woman smiles slightly after hearing the response.
Audio:
Light retro synth comedy soundtrack.
Subtle hydraulic footsteps.
Servo motors.
Industrial ambience.
Distant machinery.
Soft steam vents.
Perfect dialogue clarity.
Accurate natural lip synchronization.
Performance:
Relaxed comedic timing.
Natural facial expressions.
Subtle eye contact.
Believable body language.
No exaggerated acting.
Constraints:
Exactly one woman.
Exactly one robot.
No additional robots.
No extra foreground characters.
No costume changes.
No identity drift.
No face distortion.
No camera shake.
No duplicated subjects.
No object morphing.
No text overlays.
No subtitles.
No logos.
No watermarks.
No AI artifacts.
seedance 2.0 for AI OFM reels is f*cking insane right now
i feed it one still from nano banana pro and it generates a full reel in under 90 seconds
- natural head movement and body motion
- realistic lip sync for voice note content
- consistent face across every clip
- physics that don't break (hair, clothing, lighting)
- looks like real phone footage not CGI
one image. one prompt. 90 seconds. done
our reels are pulling 40k+ views and nobody in the comments has clocked it once
here's what the output looks like:
comment "REELS" and i'll send you the full workflow (must be following + RT)
有朋友问我这种怎么搞的,其实很简单,Seedance 2.0 上传四张图,然后复制如下提示词,生成出来后让codex 换上经典配乐,再帮忙剪辑下效果即可,提示词还是在某音学的🐶:
8K 超清,超精细照片;场景位于高层玻璃护栏露天观景露台,黄昏蓝调时刻,远处是灯火初亮的现代化摩天都市天际线,背景半空悬浮一架黑色民用直升机,直升机灯光亮起,柔和黄昏环境光,氛围感奢华财阀聚会风,深景深效果,城市远景朦胧虚化做背景烘托,斯皮尔伯格级精致镜头美学,画面构图均衡大气,无背景音乐无字幕。
分镜 1 | 00:00-00:02 中景起幅・弧形左移
中景起幅,镜头平稳起幅后缓慢向左侧做弧形横移;女二【@图 1】超模气质的女性身着香槟米调闪钻长款晚礼服,裙摆蓬松拖地,礼服肩部点缀蓬松白色羽毛,斜身站立呈一只手轻托下颌低头沉思状态,随即抬眼看向镜头,眼神充满疏离感;前景预留失焦埋伏区域,焦点稳定锁定女二,前景自然虚化,景深层次分明;
分镜 2 | 00:02-00:06 拉焦切镜・女一【@图 2】贴脸特写,镜头持续向左弧形横移,同步执行极速前景拉焦;女一同样穿着香槟米调闪钻长款晚礼服,裙摆蓬松拖地,礼服肩部点缀蓬松白色羽毛;画面前景原本完全失焦的区域,女一转过头,以由下而上的低角度形成极度贴脸的大特写,撞满画面前景,眼神慵懒灵动,唇色粉嫩自然;背景女二低头闭眼彻底虚化甩至远端,前后景形成极强空间拉扯张力;焦点精准锁死女一面部,背景全虚;
分镜 3 | 00:06-00:12 斜向后撤升起・镜头不间断执行大范围斜向后撤 + 升起的摇臂运动,画面空间逐级拓展;视野随镜头运动呈倍数级放大,女一从侧转身正对镜头同时双手交叉抱在胸前,站在画面中央,掌控全局的霸气,女三【@图 3】倚靠露台栏杆站在画面最左侧,女四【@图 4】单手叉腰站立在画面左侧第二位,女二侧身回眸站在女一右侧,女三和女四也穿着香槟米调闪钻长款晚礼服;随镜头后撤逐步拉远,半空悬浮一架黑色民用直升机,直升机灯光亮起;
分镜 4 | 00:12-00:15 大全景落幅 权力定格
大全景落幅,镜头完全拉远后
视频时长:15秒
🤖 Made with AI
Cukong Crypto retweeted
Created with Seedance 2.0 @budgetpixel
15-Second Video Prompt
Style: Ultra-realistic, cinematic winter documentary, 4K HDR, natural lighting, shallow depth of field, smooth camera movement, soft colour grading, peaceful atmosphere. Maintain the same woman throughout every scene (mid-20s, natural appearance, long dark brown hair tied in a loose ponytail, cream knitted sweater, dark leggings, wool socks, no makeup-heavy look).
Scene 1 (0–3s)
Aerial drone shot of a small wooden cabin hidden deep inside a snow-covered pine forest. Heavy snow falls gently while warm golden lights glow from the windows. Thin smoke rises from the chimney as the camera slowly descends toward the cabin.
Scene 2 (3–6s)
Inside the cabin, the same woman sits beside a crackling stone fireplace wrapped in a soft wool blanket, reading a book. Steam rises naturally from a ceramic mug of hot chocolate on a rustic wooden table. Firelight flickers across the room, creating a warm contrast with the cold blue light outside.
Scene 3 (6–10s)
She opens the wooden front door and steps onto the snowy porch. Snowflakes land softly on her sweater and hair. She smiles, takes a slow deep breath, and looks across the silent forest while the camera gently circles around her.
Scene 4 (10–13s)
Close-up cinematic shots: boots leaving fresh footprints in untouched snow, her gloved hand brushing snow from pine branches, snowflakes settling on the warm cabin window, and smoke drifting into the crisp winter air.
Scene 5 (13–15s)
Wide cinematic pullback from the cabin at dusk. Snow continues falling as the glowing cabin becomes smaller among the vast white forest. The scene ends with a calm, peaceful atmosphere and the feeling of complete solitude.
Camera
Slow drone movements
Smooth gimbal tracking
Gentle push-ins
Macro close-ups
No fast cuts or shaky movement
Audio
Gentle wind
Soft snowfall ambience
Fireplace crackling
Distant birds
Quiet piano or acoustic instrumental
Colour Grade
Warm amber tones indoors, cool blue-white winter tones outdoors, creating a cosy contrast.
End Text
Sometimes the quietest places leave the strongest memories.
🤝 Paid partnership · 🤖 Made with AI
EVERYONE PROMPTS THE ACTION. ALMOST NOBODY LOCKS THE IDENTITY — WHICH IS WHY TWO-CHARACTER SCENES FALL APART.
Two freerunners racing across Tokyo rooftops, eight cuts, corkscrews over a rooftop gap at the end. The parkour is the easy part. Keeping them two separate people who never blend into each other is the part that actually breaks.
Here's the full prompt built that way. Attach two reference photos as image_1 and image_2, and the same structure works for any multi-character action piece:
FORMAT: 15 seconds, 16:9, 1080p, 8-cut cinematic ultra-advanced parkour footage.
CHARACTERS: Two realistic individuals from image_1 and image_2. Use the attached images as absolute character references, and fully maintain the facial features, hairstyles, hair colors, skin textures, body types, height differences, outfits, color schemes, and age appearances of each person across all cuts. No altering into different people, face swaps, outfit changes, hairstyle changes, or mixing of the two individuals' features.
SETTING: A sunny modern Japanese city reminiscent of Tokyo, Shibuya, and Yokohama — rooftops, alleys, staircases, railings, pipes, concrete walls. The two protagonists, as equals, race through at high speed running side by side, following, crossing paths, and coordinating.
CUTS:
1. (00:00–00:01.60) Low-angle rear tracking. Both accelerate side by side and simultaneously kong vault over separate obstacles.
2. (00:01.60–00:03.40) Front low-angle. One wall runs the left wall, the other the right, then tic-tac to cross in midair and land on opposite rooftops.
3. (00:03.40–00:05.20) Lateral tracking. Consecutive precision jumps, then cat leaps to grab and climb a high wall.
4. (00:05.20–00:07.20) Rooftop tracking. The leader dash vaults, the trailer websters over the gap, then they swap front and back positions.
5. (00:07.20–00:09.20) Overhead moving camera. Both dive roll, then run side by side to speed vault a long railing.
6. (00:09.20–00:11.30) Handheld retreating from the front. One underbars, the other side flips, conquering the obstacle simultaneously.
7. (00:11.30–00:13.20) Drone from diagonal rear above. Both palm spin off left and right walls, kong vault, accelerate into the final jump.
8. (00:13.20–00:15.00) Climax. Both leap a large rooftop gap, each doing a corkscrew, camera circling them in midair as they land on separate rooftop edges — then run side by side into the distance.
QUALITY: Live-action film quality. World-championship-level smooth freerunning. Realistic center-of-gravity shifts, muscle movement, natural landing impacts, swaying hair and clothing. Sharp background, natural motion blur only during high-speed movement.
PROHIBITED: Facial distortion, altering into different people, face or body swaps, outfit changes, hairstyle changes, body type changes, limb multiplication, duplicates, body fusion, penetration, warping, floating, unnatural landings, anime style, CG style.
A few things worth noticing about why it's built this way:
The character block does identity work three separate times — the reference images, the "fully maintain" list, and the prohibited list at the end. That redundancy isn't padding; each one closes a different door the model tends to walk through.
The prohibited list names the exact failure modes — face swaps, body fusion, limb multiplication. Telling the model what not to do is more effective here than describing what you want, because these are the specific ways two-character scenes collapse.
Every cut assigns each person a distinct action — one wall runs left, the other right; one underbars, the other side flips. Giving them separate roles keeps them functionally two people, so the model can't average them into one.
And the cuts are individually timed and framed. Long continuous motion is where identity drift creeps in — breaking it into eight discrete shots gives the model less room to blend them.
Made in Seedance 2.0.
Cukong Crypto retweeted
These shots are 100% AI-made.
Created with Seedance 2.0 + GPT Image 2.
No reference images. No animators. Just prompts
Prompt below:
Cukong Crypto retweeted
Nothing can match the realism of Seedance 2.5
Prompt:
**CAMERA:**
DV 16mm tape camcorder handheld feel. POV of CHASE holding the camera herself throughout each location, occasionally propped briefly for hands-free moments. Hand shake, misaligned framing, delayed focus pulls, clumsy zooms, occasional face cut-off framing, imperfect shots. Camcorder never appears on screen.
**LOOK:**
Soft, slightly blurry tape quality, faint tape noise, bloomed highlights, flickering auto-exposure, muted contrast, realistic skin tones — lighting shifts per location (dim early-morning dorm → fluorescent vocal room → mirror-lit practice room → stark evaluation room lighting).
**STYLE:**
Nostalgic, determined montage feel — younger, rawer energy than her current polished vlogs, evoking a flashback to trainee days. Quick cuts stitched across locations, reflective voiceover narration plays over the visuals instead of synced dialogue. Mood shifts from tired to focused to nervous, ending on quiet resolve.
**Character**
CHASE — Korean trainee (ref image), late teens/early 20s (younger, flashback version of herself). Long straight black hair, often in a simple ponytail or braid, minimal to no makeup, tired but determined eyes. Wearing simple practice clothes — a plain long-sleeve top and joggers or sweats, fully covering arms and torso, no jewelry.
**Setting Progression**
Trainee dorm room (early morning) → vocal lesson room (morning) → dance practice room (afternoon) → evaluation room (evening).
**Storyboard (15s, 5 cuts)**
1. *(~3s, dorm, propped camera, dim early light)* She wakes up before sunrise, rubbing her eyes, quietly getting ready in the dark so as not to wake roommates. VOICEOVER (CHASE): "Back then, mornings started before the sun even came up."
2. *(~3s, vocal room, handheld, fluorescent light)* She stands at a mic, sheet in hand, focused and a little nervous, mouthing through warmups. VOICEOVER (CHASE): "Every note mattered — you never knew who was listening."
3. *(~3s, practice room, handheld, mirror wall)* She's mid-motion rehearsing choreography, sweat forming, repeating a step over and over. VOICEOVER (CHASE): "I must have run that same eight-count a hundred times."
4. *(~3s, practice room, macro insert)* Close-up on her feet repeating the same step, breathing audible, tape blur softening the motion. No narration — ambient practice room sound only.
5. *(~3s, evaluation room, wide-to-close, tense finish)* She stands nervously in front of unseen evaluators, hands clasped, taking a steadying breath before the camera holds on her face. VOICEOVER (CHASE): "That was the version of me that started all of this."























