Pinned Tweet
AI photo edits keep changing things you NEVER asked them to.
Fix the light → new face.
Remove a stranger → rebuilt street.
Retouch skin → plastic doll.
The fix: tell AI what to KEEP.
10 copy-paste prompts for GPT Image 2.5 + Nano Banana 2.1 ↓
Stop generating 9 random AI portraits.
Build ONE world. Shoot it 9 ways.
I made a cozy café photo dump with the same girl, sweater, café, and light—then changed the moments.
It feels like one real afternoon, not 9 unrelated AI images.
Copy-paste prompt + 9-shot plan ↓
🤖 Made with AI
Prompt:
THE VISUAL SYSTEM
LOCK:
Same face, hairstyle, and natural skin texture
Same cream knitted sweater and accessories
Same cozy café, bookshelves, plants, and wooden furniture
Same warm window light and natural color palette
Same laptop, iced latte, notebook, and pen
CHANGE:
Facial expression
Pose and body language
Camera angle and distance
Interaction with everyday objects
The little moments that tell a story
COPY-PASTE MASTER PROMPT:
Create ONE ultra-realistic vertical 9:16 lifestyle photograph as part of a cohesive editorial photography series titled "Café Diaries — A Slow Afternoon."
IDENTITY LOCK
Use the uploaded reference photograph as the primary identity reference. Preserve the exact recognizable facial structure, eye shape and color, natural skin tone, facial proportions, and long naturally wavy dark brown hair.
Maintain the same young adult woman across every photograph. No identity drift, facial alterations, or artificial beautification.
WARDROBE LOCK
Same oversized cream chunky-knit sweater, delicate gold earrings, minimal accessories, and effortless everyday makeup.
ENVIRONMENT LOCK
All photographs take place in the same cozy café on a peaceful, sunny afternoon.
Keep consistent:
Warm wooden bookshelves
Hanging green plants
Large sunlit window
Framed pictures
Glowing table lamp
Wooden table and comfortable seating
LIFESTYLE PROPS
A silver laptop, clear iced latte with a black straw, open notebook, and dark pen.
Show these objects only when naturally visible from the selected camera angle. Maintain spatial continuity.
PHOTOGRAPHIC STYLE
Authentic, photorealistic lifestyle editorial photography.
Soft natural daylight mixed with subtle warm café lighting. Gentle highlights, realistic shadows, warm organic tones, shallow but believable depth of field.
Natural skin pores, delicate facial details, individual hair strands, and richly textured knitted fabric.
Candid expressions, relaxed body language, spontaneous moments, slightly imperfect framing.
The result should feel like photographs captured by a friend during one beautiful afternoon, not a polished commercial photoshoot.
SCENE:
[INSERT ONE SHOT DESCRIPTION BELOW]
STRICT OUTPUT RULES
Generate exactly ONE standalone photograph per prompt.
Vertical 9:16. No collage, no grid, no contact sheet.
Reuse the same identity and environment references for every generation.
Preserve identity, outfit, café setting, and photographic style while changing only the individual moment and composition.
NEGATIVE PROMPT
Identity drift, different face, changing wardrobe, inconsistent locations, plastic skin, excessive skin smoothing, distorted hands, extra fingers, artificial poses, oversaturated colors, harsh HDR, CGI appearance, commercial studio lighting, duplicate subjects, text, logos, watermarks.
THE 9-SHOT PLAN
01 — THE COZY PORTRAIT
Resting her cheek gently on one hand, making direct eye contact with a soft smile. Intimate eye-level portrait.
02 — DEEP IN WORK
Typing naturally on her silver laptop, eyes focused downward. Candid three-quarter angle with sunlight illuminating her hair.
03 — COFFEE BREAK
Holding her iced latte, casually sipping through the straw while glancing away from the camera. Relaxed medium close-up.
04 — LITTLE NOTES
Writing thoughtfully in her notebook with the laptop nearby. Slightly elevated camera angle, emphasizing her hands and the wooden tabletop.
05 — LOST IN THOUGHT
Resting her chin on her hand while looking through the café window. Soft side profile with beautiful natural highlights.
06 — CANDID LAUGH
Laughing genuinely at something off-camera. Relaxed shoulders, slightly turned head, spontaneous expression.
07 — SMALL DETAILS
Close-up of her hand holding the iced latte beside her notebook and laptop. Focus on condensation, knit texture, and beautiful natural light.
08 — ONE MORE CHAPTER
Quietly reading an open book. Three-quarter portrait, soft concentration, cozy bookshelves in the background.
09 — AFTERNOON GLOW
Looking back over her shoulder toward the camera with a gentle smile. Golden afternoon sunlight catching her hair.
🤖 Made with AI
HOW TO USE
Upload your reference portrait.
Copy the Master Prompt.
Replace [SCENE] with one of the 9 descriptions.
Generate each image separately using the same reference.
Repeat until you've built your complete series.
One person. One place. Nine little moments.
That's how you make an AI photo dump feel like a real memory.
Save this for your next AI photoshoot. ✨
Which frame would you recreate first — 1 to 9?
🎃🐺🧙♀️Bro, this is NOT how Halloween was supposed to go
Prompt⬇️
14-SECOND VIDEO PROMPT — DRIVE-THRU FROM HELL
STYLE: Photoreal raw flash footage, like a lost Halloween night tape from 2003. Cheap early-2000s consumer camcorder, extreme fisheye lens, strong barrel distortion, harsh on-camera light, blown highlights, hard shadows, near-black backgrounds.
Heavy digital grain, slight interlacing, smeared motion blur, green-tinted colors, auto-exposure pumping, and occasional focus hunting.
Chaotic documentary realism: a real camera operator panicking inside a car, frantic handheld whip-pans, crash zooms, accidental bumps, crooked framing, and monsters rushing directly into the fisheye lens. Absurd meme energy played completely seriously.
FAST PACING: Hard jump cuts every 0.5–1 second. No transitions, cinematic polish, slow motion, text, logos, or watermarks.
LOCATION: Run-down American fast-food drive-thru at 2 AM. Old sedan, glowing menu board, pickup window, flickering fluorescent lights, wet asphalt, nearly black parking lot.
CAST — KEEP IDENTICAL THROUGHOUT:
Devil woman: long dark hair, red horns, crimson costume, sheer sleeves, red tulle skirt. Driver.
Dale: enormous grey-brown werewolf, yellow eyes, sharp teeth, torn red varsity jacket.
Three skeletons: aviator sunglasses, black tactical vests; one wears an orange cap and white sneakers.
Elderly witch: wrinkled face, wild grey hair, pointed hat, black feathered cape, riding a battered lawnmower.
Headless Horseman: burgundy coat, truly headless, riding a black horse, holding a flaming pumpkin.
Living jack-o’-lantern: small orange pumpkin with glowing carved eyes and mouth.
14-SECOND SEQUENCE — 18 RAPID SHOTS:
01 (0–0.8s): Flash-lit fisheye close-up. Devil woman bored at the drive-thru menu.
02 (0.8–1.6s): Giant werewolf suddenly lunges beside the menu, snarling into the lens.
03 (1.6–2.4s): Whip-pan inside car. Werewolf crammed into back seat, stealing fries.
04 (2.4–3.2s): Three skeleton employees suddenly crowd the pickup window.
05 (3.2–4s): Orange-cap skeleton crushes a drink tray; soda sprays toward camera.
06 (4–4.8s): Witch races past the car on a sputtering lawnmower, cape flying.
07 (4.8–5.6s): Lawnmower hits a trash can; cups and wrappers explode through the flash.
08 (5.6–6.4s): Crash zoom. Devil woman stares at camera in disbelief; werewolf eats fries behind her.
09 (6.4–7.2s): Loud THUD on car roof. Camera jerks upward toward a moving shadow.
10 (7.2–8s): Headless Horseman gallops past the driver-side window, raising a flaming pumpkin.
11 (8–8.8s): Black horse thrusts its giant snout inches into the fisheye lens, snorting.
12 (8.8–9.6s): Inside car. Devil woman notices a mysterious takeout bag moving by itself.
13 (9.6–10.4s): Extreme close-up inside bag. Strange orange glow flickers.
14 (10.4–11.2s): Living jack-o’-lantern bursts from the bag, shrieking into the lens.
15 (11.2–12s): Devil woman screams; werewolf yelps; skeletons recoil in terror.
16 (12–12.8s): Total panic. Witch reverses, skeletons scatter, horse backs away, werewolf flees.
17 (12.8–13.4s): Orange-cap skeleton crashes into the operator. Camera tumbles onto wet asphalt.
18 (13.4–14s): Fallen sideways camera. Living pumpkin rolls straight toward the fisheye lens, opens its mouth impossibly wide, and SHRIEKS. HARD CUT TO BLACK.
CAMERA: Aggressive fisheye distortion, brutal direct flash, frantic human-operated handheld motion, violent whip-pans, unstable framing, realistic collisions, exposure flicker, and heavy motion blur. Subjects must look physically present, not CGI. Maintain recognizable faces, consistent costumes, believable anatomy, and exactly three skeletons. No duplicated characters or extra limbs.
AUDIO: Raw distorted camcorder microphone only. Drive-thru static, engine rumble, werewolf snarls, clattering bones, splashing soda, lawnmower revving, horse hooves, panicked screams, rustling paper bag, pumpkin shriek, and camera hitting asphalt.
🤖 Made with AI
📼 HOW I MADE IT — FULL WORKFLOW
1. Lock the visual style
Photoreal raw flash footage, like a lost Halloween night tape from 2003. Cheap consumer camcorder, extreme fisheye barrel distortion, harsh on-camera flash-style video light, blown highlights, hard shadows, near-black backgrounds, heavy digital grain, green-tinted colors, interlacing, smeared motion blur, auto-exposure pumping, and occasional focus hunting.
Frantic handheld whip-pans, crash zooms, camera collisions, crooked framing, and subjects rushing straight into the lens.
Absurd horror-comedy played completely seriously.
2. Build the 18-shot storyboard
Use the same characters and location throughout:
Devil woman driving an old sedan.
Giant werewolf in a torn red varsity jacket.
Three skeleton employees wearing tactical vests and aviators.
Elderly witch riding a lawnmower.
Headless Horseman on a black horse.
Living jack-o’-lantern hidden inside a takeout bag.
3. Animate the sequence
0–2.4s: Late-night order, werewolf jumpscare, stolen fries.
2.4–4.8s: Skeleton employees crowd the window, soda explodes, witch arrives.
4.8–7.2s: Trash-can collision, devil woman’s reaction, impact on the car roof.
7.2–9.6s: Headless Horseman arrives, horse close-up, mysterious takeout bag.
9.6–12.0s: Orange glow, living pumpkin jumpscare, everyone panics.
12.0–14.0s: Monsters flee, camera falls, pumpkin delivers the final scare.
4. Camera and editing rules
Hard jump cuts every 0.5–1 second. No smooth transitions, no slow motion, no polished CGI, no cinematic camera movement.
Make every shot feel like a real person accidentally captured the chaos.
5. Sound design
Raw camcorder microphone only: drive-thru static, werewolf growls, bones clattering, soda splashing, lawnmower engine, horse hooves, screams, paper bag rustling, pumpkin shrieking, and camera impacts.
No music. No score. No voiceover.
FINAL PROMPT RULE:
14 seconds. 18 shots. Extreme fisheye. Brutal direct flash. Heavy camcorder grain. Chaotic documentary realism. Consistent characters. Physical interactions. Hard cuts only. No text, logos, or watermarks.
The goal isn’t cinematic perfection.
It’s making AI footage look like a completely unhinged Halloween tape someone actually found from 2003. 🎃
Chicken Night🍗🍟🍔
I'm sure you're gonna want to order fried chicken after seeing this.
Prompt⬇️
TITLE: CRISPY FRIED CHICKEN NIGHT
DURATION: 15 seconds
FORMAT: Vertical 9:16, authentic iPhone selfie-vlog footage.
STYLE
Photorealistic casual UGC food vlog, shot on a handheld iPhone. Natural TikTok/Reels aesthetic, warm apartment lighting, spontaneous personality, quick jump cuts, imperfect handheld framing, realistic smartphone autofocus breathing, subtle sensor grain and genuine room audio. NOT a polished food advertisement.
CHARACTER — SKAI
One consistent young adult woman in her early 20s, friendly and playful lifestyle-vlogger personality. Shoulder-length dark brown hair loosely tied back, natural makeup, oversized cream T-shirt, small silver hoop earrings. Keep her face, voice, hairstyle, clothing and appearance identical across every shot. If a character reference is supplied, preserve that identity instead.
FRIED CHICKEN — PRODUCT CONSISTENCY
A small unbranded kraft-paper takeaway box containing three pieces of classic Southern-style crispy fried chicken, with a separate serving of golden fries. The hero piece is a large bone-in drumstick with a distinctive irregular, craggy golden-brown crust, tiny crunchy flakes, subtle natural oil sheen and realistic proportions. Maintain exactly the same drumstick, coating, color, box and food arrangement throughout. After the bite, preserve a realistic bite mark. If an actual fried-chicken reference image is provided, use that image as the primary product reference. Never transform the chicken into a sandwich or boneless nugget.
SHOT 1 (0–2s) — SELFIE HOOK
Front-facing iPhone selfie. Skai smiles casually while holding the closed fried-chicken takeaway box beside her face.
Dialogue, excited but natural: “Fried chicken night!”
Slight camera shake, warm home ambience.
SHOT 2 (2–4s) — UNBOXING
Quick jump cut to a handheld downward phone angle. Skai opens the cardboard box, revealing three golden crispy chicken pieces. Realistic cardboard rustling and subtle hand movement. Slightly imperfect framing.
SHOT 3 (4–6s) — CRISPY DETAIL
Abrupt handheld push-in toward the hero drumstick. Brief natural autofocus adjustment. Clearly reveal irregular crunchy coating, flaky golden ridges and subtle oil sheen. Real food texture, never CGI or exaggerated artificial crunch.
SHOT 4 (6–8s) — THE PICKUP
Close handheld shot as Skai lifts the same drumstick from the box. Fingers gently press the crispy crust. A couple of tiny crumbs naturally fall. Subtle real crust crackle. Correct fingers, natural grip, realistic weight and movement.
SHOT 5 (8–10s) — FIRST BITE
Jump cut back to selfie perspective. Skai takes one genuine bite from the drumstick. An audible natural CRUNCH is synchronized exactly with her teeth breaking the crust. She pauses with surprised, delighted eyes, briefly finishes the bite and says: “Wait... SO good.” Authentic reaction, not exaggerated acting.
SHOT 6 (10–12s) — CASUAL FOOD B-ROLL
Quick close-up of the same bitten drumstick above the open box. A small bite mark reveals naturally juicy, fully cooked chicken meat under the crispy crust. Skai casually reaches for a fry with her free hand. Handheld refocusing, quiet room ambience and light packaging rustle.
SHOT 7 (12–14s) — FRIENDLY RECOMMENDATION
Front-facing selfie again. Skai playfully raises the bitten drumstick toward the camera like a casual toast. She smiles directly into the lens.
Dialogue: “You need this.”
Slight camera wobble and spontaneous laughter.
SHOT 8 (14–15s) — FREEZE FRAME
Freeze on the genuine final moment with the drumstick close to the camera and Skai smiling in the background.
Single clean on-screen overlay:
“crispy cravings = solved 🍗”
No other titles, logos or watermarks.
CAMERA & EDITING
One coherent evening in the same apartment. Fast natural jump cuts, realistic iPhone front/rear-camera behavior, slight exposure shifts, occasional imperfect focus, close handheld food details. No cinematic camera movements, no professional food styling, no artificial slow motion, no stabilized commercial footage.
AUDIO
Natural young female voice with consistent tone and realistic lip synchronization. Subtle apartment ambience, packaging rustle, gentle crust crackle, one clearly synchronized satisfying bite crunch. No exaggerated cartoon crunch, dramatic music or artificial sound design. Keep dialogue intelligible.
NEGATIVE PROMPT
Commercial production, studio lighting, cinematic color grading, beauty filters, CGI food, plastic fried chicken, smooth artificial coating, excessive dripping oil, cheese pulls, warped hands, extra fingers, changing chicken pieces, inconsistent bite marks, duplicate food, distorted mouth, unnatural chewing, broken anatomy, perfect stabilization, scripted acting, text glitches, unwanted captions, fake logos.
PRIORITY
Authenticity first. Food consistency second. Audible crispy bite third. Fast, relatable, spontaneous social-media energy throughout.
@grok @imagine
🤖 Made with AI
POV: you just wanted fries at 2 AM. 🍟💀
The drive-thru employees are skeletons.
A werewolf is stealing your fries.
A witch cuts the line on a lawnmower.
And the Headless Horseman shows up.
Then you open your takeout bag… 🎃
Halloween is coming.
14 seconds. 18 shots. Pure 2003 camcorder chaos.
Full prompt ↓
🤖 Made with AI
🎃👻Bro, this is NOT how Halloween was supposed to go
Prompt:
THE CURSED CONVENIENCE STORE
14-Second Chaotic Halloween Found-Footage Video
STYLE — STRICT VISUAL LOCK
Photoreal raw flash footage, like a lost Halloween night tape from 2003.
Shot on a cheap early-2000s consumer camcorder with an extreme fisheye lens: strong barrel distortion, harsh on-camera flash-style video light blasting anything close to the lens with blown-out highlights and hard shadows. Everything beyond a few meters falls into an underexposed, nearly black background.
Heavy camcorder noise, ugly digital grain, subtle interlacing, smeared motion blur, auto-exposure pumping, occasional focus hunting, slightly crushed blacks, green-tinted digital colors, blown highlights, and imperfect low-resolution recording.
Chaotic documentary realism. The camera operator is a real person physically running, stumbling, ducking, and reacting to everything happening around them.
Frantic handheld whip-pans, sudden crash zooms, violent camera bumps, crooked framing, partial subject cutoffs, lens smears, accidental close-ups, and characters repeatedly shoved directly into the fisheye lens.
Everything feels accidentally captured, never choreographed or professionally filmed.
Meme energy: completely absurd situations played with absolute seriousness.
No cinematic lighting. No polished CGI appearance. No stabilized camera. No smooth tracking shots. No slow motion.
FAST PACE: Aggressive hard jump cuts every 0.5–1 second. No transitions, no fades, no dissolves.
No on-screen text, no timestamp overlay, no subtitles, no titles, no logos, no watermark.
LOCATION — THE CURSED CONVENIENCE STORE
A small, run-down American roadside convenience store at 2 AM on Halloween night.
Old fluorescent ceiling lights flicker weakly. Narrow aisles are packed with colorful chips, candy, soda bottles, cheap Halloween decorations, and stacked cardboard displays.
An old cash register sits beside a scratched checkout counter. Refrigerated drink coolers line the back wall. Large automatic glass doors face a nearly pitch-black parking lot.
The entire store feels cramped, dirty, cluttered, and realistically lived-in.
The camera's harsh mounted light is the dominant illumination. Nearby faces, costumes, merchandise, and objects are violently overexposed, while the deeper aisles disappear into dark greenish shadows.
All events happen inside this same store or directly at its entrance.
Maintain a consistent physical layout across every shot.
CAST — STRICT CHARACTER CONSISTENCY
THE DEVIL WOMAN — MAIN CHARACTER
A striking adult woman with long dark hair, small pointed red devil horns, a dramatic crimson-red costume with a layered red tulle skirt, sheer red sleeves, fitted bodice, and black high heels.
She looks bored, irritated, and completely unimpressed by the supernatural chaos.
She is working the overnight checkout shift.
If a character reference image is provided, preserve her exact facial identity and recognizable appearance.
Keep her face, hairstyle, horns, costume, and accessories identical throughout every shot.
DALE THE WEREWOLF
An enormous hulking grey-brown werewolf with matted shaggy fur, glowing yellow eyes, massive clawed hands, and an elongated snarling snout full of sharp teeth.
He wears a torn red varsity-style bowling jacket with shredded sleeves.
He looks terrifying but behaves like an oversized, badly trained dog.
THE THREE SKELETONS
Three full-body human skeletons with realistic bleached off-white bones.
All wear mirrored aviator sunglasses and black tactical vests.
Skeleton A wears a bright orange baseball cap and white sneakers.
Skeletons B and C wear no hats.
They move like reckless, hyperactive troublemakers. Their bones rattle and clatter with every sudden movement.
Their sunglasses, vests, and individual accessories must remain consistent.
THE ELDERLY WITCH
A frail, deeply wrinkled elderly woman with wild grey hair, sharp black cat-eye glasses, a crooked grin, a pointed black witch hat, and a long black feathered cape.
She drives a battered, sputtering ride-on lawnmower at ridiculous speed.
THE HEADLESS HORSEMAN
A tall, imposing headless rider wearing a dark burgundy velvet coat and black leather gloves.
He has absolutely no head, face, or neck above the coat collar.
He rides a powerful black horse with a thick flowing mane and carries a glowing carved jack-o'-lantern.
The horse is physically present in the store, interacting with shelves, doors, and the cramped environment.
THE TINY GHOST — FINAL REVEAL
A ridiculously small ghost, barely knee-high, resembling a floating wrinkled white bedsheet with two uneven black eyeholes.
No visible body, feet, or hands.
No magical glow, no particles, no elaborate CGI effects.
It looks harmless, almost comically cheap, like a real Halloween costume somehow floating above the floor.
Despite its tiny size, every other monster is inexplicably terrified of it.
SHOT-BY-SHOT SEQUENCE — EXACTLY 14 SECONDS
SHOT 01 — 0.0–0.8s | THE NIGHT SHIFT
Extreme fisheye close-up behind the convenience-store counter.
The devil woman leans against the register, bored, chewing gum and staring directly into the lens.
Harsh flash flattens her face. The aisles behind her are nearly black.
A faint electrical buzz fills the room.
SHOT 02 — 0.8–1.6s | WEREWOLF AMBUSH
BANG!
The automatic glass doors slam open.
Dale the enormous werewolf charges out of the darkness, sprinting directly toward the camera.
His snarling face suddenly fills the entire fisheye frame.
The operator screams and stumbles backward.
SHOT 03 — 1.6–2.4s | DOG TREAT ROBBERY
Violent whip-pan.
Dale completely ignores the devil woman.
Instead, he grabs a huge bag of dog treats from a shelf, tears it open with his teeth, and shoves his massive furry snout toward the lens.
Crumbs fly everywhere.
The camera struggles to focus.
SHOT 04 — 2.4–3.2s | SKELETON RAID
Hard jump cut.
Three skeletons suddenly burst through the entrance and leap onto the checkout counter.
Their mirrored sunglasses catch the harsh flash.
The skeleton in the orange baseball cap lands directly in front of the camera.
Bones clatter loudly against the register.
SHOT 05 — 3.2–4.0s | SODA DISASTER
Extreme fisheye close-up.
The orange-cap skeleton violently shakes a large soda bottle, then twists off the cap.
The other skeletons throw their arms into the air.
SHOT 06 — 4.0–4.8s | LENS ATTACK
The soda erupts directly toward the camera.
Sticky liquid splashes across the fisheye lens.
The flash catches thousands of droplets.
The operator yells "YO!"
The camera jerks sideways, struggling to see through the smear.
SHOT 07 — 4.8–5.6s | WITCH ON WHEELS
A deafening lawnmower engine suddenly roars.
The elderly witch tears down a narrow snack aisle on her battered ride-on lawnmower.
Her feathered cape whips behind her.
She passes inches from the fisheye lens, grinning maniacally.
SHOT 08 — 5.6–6.4s | SNACK AISLE COLLISION
The lawnmower smashes into a towering cardboard display of potato chips.
The display collapses.
Hundreds of colorful snack bags fly toward the camera.
The operator ducks beneath falling merchandise.
SHOT 09 — 6.4–7.2s | DEVIL WOMAN REACTS
Crash zoom onto the devil woman.
She ducks behind the counter as bags rain down around her.
She looks straight into the camera with an expression of complete disbelief.
A potato chip bag slaps against the lens.
SHOT 10 — 7.2–8.0s | THE HEADLESS CUSTOMER
The automatic doors violently slide open again.
The Headless Horseman rides his enormous black horse directly into the convenience store, barely squeezing through the entrance.
His burgundy coat flaps.
He raises a glowing jack-o'-lantern.
The camera whips backward in panic.
SHOT 11 — 8.0–8.8s | HORSE FACE CLOSE-UP
The horse thrusts its enormous nostrils straight into the fisheye lens.
Extreme barrel distortion makes the snout look absurdly huge.
The horse snorts violently, fogging the lens.
The camera operator stumbles into a merchandise shelf.
SHOT 12 — 8.8–9.6s | SOMETHING IN THE COOLER
A strange metallic CLUNK comes from the refrigerated drink section.
The camera snaps toward the back wall.
One cooler door slowly swings open.
A tiny white shape is barely visible in the darkness near the floor.
For a split second, the chaos goes strangely quiet.
SHOT 13 — 9.6–10.4s | THE TINY GHOST
The tiny white-sheet ghost slowly floats out of the cooler.
It is barely knee-high.
Its two uneven black eyeholes face the camera.
The harsh flash makes its white sheet glow with ugly overexposure.
It tilts slightly to one side.
SHOT 14 — 10.4–11.2s | BOO
Smash cut to an extreme fisheye close-up of the tiny ghost.
It suddenly lunges a few inches toward the lens.
In a ridiculously small, squeaky voice, it whispers:
"BOO."
SHOT 15 — 11.2–12.0s | MONSTERS TERRIFIED
Violent whip-pan.
Dale the giant werewolf lets out a high-pitched terrified yelp and drops his dog treats.
All three skeletons recoil in panic, their jaws hanging open.
The devil woman's eyes go wide.
Everyone is genuinely horrified by this harmless little ghost.
SHOT 16 — 12.0–12.8s | TOTAL PANIC
Rapid handheld chaos.
The witch frantically reverses her lawnmower into a shelf.
The horse rears slightly and backs toward the automatic doors.
Dale scrambles away on all fours.
Skeletons collide with one another as they run.
The operator screams and tries to escape with them.
SHOT 17 — 12.8–13.4s | CAMERA DOWN
The orange-cap skeleton trips directly into the operator.
Its bones collapse into a chaotic pile.
The camera is knocked out of the operator's hands.
It tumbles toward the sticky convenience-store floor.
The picture spins violently, flash streaking across scattered bones and candy wrappers.
SHOT 18 — 13.4–14.0s | FINAL PAYOFF
The camera lies sideways on the floor.
The store behind it has gone almost completely black.
The tiny white ghost quietly floats into view.
It leans directly into the fisheye lens until its two crooked black eyeholes fill the entire frame.
One final tiny squeaky "Boo."
HARD CUT TO BLACK at exactly 14.0 seconds.
CAMERA BEHAVIOR — MANDATORY
The footage must feel like a single real camcorder recording, aggressively jump-cut into 18 frantic fragments.
The camera operator is physically present and reacts instinctively.
Camera movements should feel uncontrolled and human:
Extremely close fisheye framing.
Violent handheld shakes and whip-pans.
Sudden crash zooms.
Constant reframing as subjects move unpredictably.
People and creatures repeatedly colliding with or rushing toward the lens.
Accidental tilted horizons and imperfect framing.
Auto-exposure blowing out bright costumes and nearby faces.
Brief focus hunting after sudden movement.
Realistic motion smears during fast camera movement.
Abrupt hard cuts with no artificial transitions.
Every event must feel captured in the moment, never staged.
Maintain real physical collisions, object weight, believable impacts, messy inertia, and tangible environmental interaction.
The monsters should look physically present inside the store, not composited into the footage.
AUDIO — RAW CAMCORDER MICROPHONE ONLY
No music. No soundtrack. No cinematic score.
Only harsh, distorted, clipping early-2000s camcorder microphone audio.
Include:
Weak fluorescent electrical buzzing.
Automatic doors rattling and slamming.
Werewolf growls, panting, and a terrified yelp.
Claws scraping against the tile floor.
Dog treat bags ripping and crunching.
Skeleton bones clattering against the counter.
Soda bottle popping and liquid spraying.
People shouting "YO!" and screaming in panic.
A sputtering ride-on lawnmower engine.
Shelves collapsing and snack bags scattering.
Heavy horse hooves clopping on tile.
Loud horse snorting.
The refrigerator compressor humming.
The tiny ghost's unexpectedly squeaky "BOO."
The camera operator's panicked breathing.
The camera hitting the floor.
Explosive sounds should briefly overload the microphone.
Voices may clip, overlap, and distort naturally.
No voiceover or narration.
FINAL CREATIVE DIRECTION
The result should feel like someone found a completely unhinged Halloween camcorder tape recorded inside a cursed American convenience store in 2003.
Ugly, loud, frantic, overexposed, dirty, ridiculous, physically chaotic, and unintentionally hilarious.
The monsters must be genuinely intimidating in appearance but absurdly incompetent in behavior.
Never turn this into a polished cinematic horror trailer, a glossy commercial, a stylized music video, or a clean CGI animation.
The humor comes from the contrast between documentary-level realism and completely ridiculous supernatural behavior.
The final joke must land clearly: a terrifying collection of enormous Halloween monsters is completely defeated by one tiny, harmless-looking ghost.
14 seconds. 18 rapid shots. Extremely aggressive fisheye. Brutal direct flash. Chaotic found-footage realism. Hard cuts only. No music. No text.
@grok @imagine #Grok
Prompt:
14-SECOND VIDEO PROMPT — DRIVE-THRU FROM HELL
STYLE: Photoreal raw flash footage, like a lost Halloween night tape from 2003. Cheap early-2000s consumer camcorder, extreme fisheye lens, strong barrel distortion, harsh on-camera light, blown highlights, hard shadows, near-black backgrounds.
Heavy digital grain, slight interlacing, smeared motion blur, green-tinted colors, auto-exposure pumping, and occasional focus hunting.
Chaotic documentary realism: a real camera operator panicking inside a car, frantic handheld whip-pans, crash zooms, accidental bumps, crooked framing, and monsters rushing directly into the fisheye lens. Absurd meme energy played completely seriously.
FAST PACING: Hard jump cuts every 0.5–1 second. No transitions, cinematic polish, slow motion, text, logos, or watermarks.
LOCATION: Run-down American fast-food drive-thru at 2 AM. Old sedan, glowing menu board, pickup window, flickering fluorescent lights, wet asphalt, nearly black parking lot.
CAST — KEEP IDENTICAL THROUGHOUT:
Devil woman: long dark hair, red horns, crimson costume, sheer sleeves, red tulle skirt. Driver.
Dale: enormous grey-brown werewolf, yellow eyes, sharp teeth, torn red varsity jacket.
Three skeletons: aviator sunglasses, black tactical vests; one wears an orange cap and white sneakers.
Elderly witch: wrinkled face, wild grey hair, pointed hat, black feathered cape, riding a battered lawnmower.
Headless Horseman: burgundy coat, truly headless, riding a black horse, holding a flaming pumpkin.
Living jack-o'-lantern: small orange pumpkin with glowing carved eyes and mouth.
14-SECOND SEQUENCE — 18 RAPID SHOTS:
01 (0–0.8s): Flash-lit fisheye close-up. Devil woman bored at the drive-thru menu.
02 (0.8–1.6s): Giant werewolf suddenly lunges beside the menu, snarling into the lens.
03 (1.6–2.4s): Whip-pan inside car. Werewolf crammed into back seat, stealing fries.
04 (2.4–3.2s): Three skeleton employees suddenly crowd the pickup window.
05 (3.2–4s): Orange-cap skeleton crushes a drink tray; soda sprays toward camera.
06 (4–4.8s): Witch races past the car on a sputtering lawnmower, cape flying.
07 (4.8–5.6s): Lawnmower hits a trash can; cups and wrappers explode through the flash.
08 (5.6–6.4s): Crash zoom. Devil woman stares at camera in disbelief; werewolf eats fries behind her.
09 (6.4–7.2s): Loud THUD on car roof. Camera jerks upward toward a moving shadow.
10 (7.2–8s): Headless Horseman gallops past the driver-side window, raising a flaming pumpkin.
11 (8–8.8s): Black horse thrusts its giant snout inches into the fisheye lens, snorting.
12 (8.8–9.6s): Inside car. Devil woman notices a mysterious takeout bag moving by itself.
13 (9.6–10.4s): Extreme close-up inside bag. Strange orange glow flickers.
14 (10.4–11.2s): Living jack-o'-lantern bursts from the bag, shrieking into the lens.
15 (11.2–12s): Devil woman screams; werewolf yelps; skeletons recoil in terror.
16 (12–12.8s): Total panic. Witch reverses, skeletons scatter, horse backs away, werewolf flees.
17 (12.8–13.4s): Orange-cap skeleton crashes into the operator. Camera tumbles onto wet asphalt.
18 (13.4–14s): Fallen sideways camera. Living pumpkin rolls straight toward the fisheye lens, opens its mouth impossibly wide, and SHRIEKS. HARD CUT TO BLACK.
CAMERA: Aggressive fisheye distortion, brutal direct flash, frantic human-operated handheld motion, violent whip-pans, unstable framing, realistic collisions, exposure flicker, and heavy motion blur. Subjects must look physically present, not CGI. Maintain recognizable faces, consistent costumes, believable anatomy, and exactly three skeletons. No duplicated characters or extra limbs.
AUDIO: Raw distorted camcorder microphone only. Drive-thru static, engine rumble, werewolf snarls, clattering bones, splashing soda, lawnmower revving, horse hooves, panicked screams, rustling paper bag, pumpkin shriek, and camera hitting asphalt. Loud sounds clip naturally. No music, score, or narration.
FINAL FEEL: A completely unhinged Halloween drive-thru disaster accidentally recorded in 2003. Ugly, frantic, overexposed, grainy, terrifying, and ridiculously funny. The final twist: the takeout order itself is the monster.
🎃👻Bro, this is NOT how Halloween was supposed to go
Prompt:
THE CURSED CONVENIENCE STORE
14-Second Chaotic Halloween Found-Footage Video
STYLE — STRICT VISUAL LOCK
Photoreal raw flash footage, like a lost Halloween night tape from 2003.
Shot on a cheap early-2000s consumer camcorder with an extreme fisheye lens: strong barrel distortion, harsh on-camera flash-style video light blasting anything close to the lens with blown-out highlights and hard shadows. Everything beyond a few meters falls into an underexposed, nearly black background.
Heavy camcorder noise, ugly digital grain, subtle interlacing, smeared motion blur, auto-exposure pumping, occasional focus hunting, slightly crushed blacks, green-tinted digital colors, blown highlights, and imperfect low-resolution recording.
Chaotic documentary realism. The camera operator is a real person physically running, stumbling, ducking, and reacting to everything happening around them.
Frantic handheld whip-pans, sudden crash zooms, violent camera bumps, crooked framing, partial subject cutoffs, lens smears, accidental close-ups, and characters repeatedly shoved directly into the fisheye lens.
Everything feels accidentally captured, never choreographed or professionally filmed.
Meme energy: completely absurd situations played with absolute seriousness.
No cinematic lighting. No polished CGI appearance. No stabilized camera. No smooth tracking shots. No slow motion.
FAST PACE: Aggressive hard jump cuts every 0.5–1 second. No transitions, no fades, no dissolves.
No on-screen text, no timestamp overlay, no subtitles, no titles, no logos, no watermark.
LOCATION — THE CURSED CONVENIENCE STORE
A small, run-down American roadside convenience store at 2 AM on Halloween night.
Old fluorescent ceiling lights flicker weakly. Narrow aisles are packed with colorful chips, candy, soda bottles, cheap Halloween decorations, and stacked cardboard displays.
An old cash register sits beside a scratched checkout counter. Refrigerated drink coolers line the back wall. Large automatic glass doors face a nearly pitch-black parking lot.
The entire store feels cramped, dirty, cluttered, and realistically lived-in.
The camera's harsh mounted light is the dominant illumination. Nearby faces, costumes, merchandise, and objects are violently overexposed, while the deeper aisles disappear into dark greenish shadows.
All events happen inside this same store or directly at its entrance.
Maintain a consistent physical layout across every shot.
CAST — STRICT CHARACTER CONSISTENCY
THE DEVIL WOMAN — MAIN CHARACTER
A striking adult woman with long dark hair, small pointed red devil horns, a dramatic crimson-red costume with a layered red tulle skirt, sheer red sleeves, fitted bodice, and black high heels.
She looks bored, irritated, and completely unimpressed by the supernatural chaos.
She is working the overnight checkout shift.
If a character reference image is provided, preserve her exact facial identity and recognizable appearance.
Keep her face, hairstyle, horns, costume, and accessories identical throughout every shot.
DALE THE WEREWOLF
An enormous hulking grey-brown werewolf with matted shaggy fur, glowing yellow eyes, massive clawed hands, and an elongated snarling snout full of sharp teeth.
He wears a torn red varsity-style bowling jacket with shredded sleeves.
He looks terrifying but behaves like an oversized, badly trained dog.
THE THREE SKELETONS
Three full-body human skeletons with realistic bleached off-white bones.
All wear mirrored aviator sunglasses and black tactical vests.
Skeleton A wears a bright orange baseball cap and white sneakers.
Skeletons B and C wear no hats.
They move like reckless, hyperactive troublemakers. Their bones rattle and clatter with every sudden movement.
Their sunglasses, vests, and individual accessories must remain consistent.
THE ELDERLY WITCH
A frail, deeply wrinkled elderly woman with wild grey hair, sharp black cat-eye glasses, a crooked grin, a pointed black witch hat, and a long black feathered cape.
She drives a battered, sputtering ride-on lawnmower at ridiculous speed.
THE HEADLESS HORSEMAN
A tall, imposing headless rider wearing a dark burgundy velvet coat and black leather gloves.
He has absolutely no head, face, or neck above the coat collar.
He rides a powerful black horse with a thick flowing mane and carries a glowing carved jack-o'-lantern.
The horse is physically present in the store, interacting with shelves, doors, and the cramped environment.
THE TINY GHOST — FINAL REVEAL
A ridiculously small ghost, barely knee-high, resembling a floating wrinkled white bedsheet with two uneven black eyeholes.
No visible body, feet, or hands.
No magical glow, no particles, no elaborate CGI effects.
It looks harmless, almost comically cheap, like a real Halloween costume somehow floating above the floor.
Despite its tiny size, every other monster is inexplicably terrified of it.
SHOT-BY-SHOT SEQUENCE — EXACTLY 14 SECONDS
SHOT 01 — 0.0–0.8s | THE NIGHT SHIFT
Extreme fisheye close-up behind the convenience-store counter.
The devil woman leans against the register, bored, chewing gum and staring directly into the lens.
Harsh flash flattens her face. The aisles behind her are nearly black.
A faint electrical buzz fills the room.
SHOT 02 — 0.8–1.6s | WEREWOLF AMBUSH
BANG!
The automatic glass doors slam open.
Dale the enormous werewolf charges out of the darkness, sprinting directly toward the camera.
His snarling face suddenly fills the entire fisheye frame.
The operator screams and stumbles backward.
SHOT 03 — 1.6–2.4s | DOG TREAT ROBBERY
Violent whip-pan.
Dale completely ignores the devil woman.
Instead, he grabs a huge bag of dog treats from a shelf, tears it open with his teeth, and shoves his massive furry snout toward the lens.
Crumbs fly everywhere.
The camera struggles to focus.
SHOT 04 — 2.4–3.2s | SKELETON RAID
Hard jump cut.
Three skeletons suddenly burst through the entrance and leap onto the checkout counter.
Their mirrored sunglasses catch the harsh flash.
The skeleton in the orange baseball cap lands directly in front of the camera.
Bones clatter loudly against the register.
SHOT 05 — 3.2–4.0s | SODA DISASTER
Extreme fisheye close-up.
The orange-cap skeleton violently shakes a large soda bottle, then twists off the cap.
The other skeletons throw their arms into the air.
SHOT 06 — 4.0–4.8s | LENS ATTACK
The soda erupts directly toward the camera.
Sticky liquid splashes across the fisheye lens.
The flash catches thousands of droplets.
The operator yells "YO!"
The camera jerks sideways, struggling to see through the smear.
SHOT 07 — 4.8–5.6s | WITCH ON WHEELS
A deafening lawnmower engine suddenly roars.
The elderly witch tears down a narrow snack aisle on her battered ride-on lawnmower.
Her feathered cape whips behind her.
She passes inches from the fisheye lens, grinning maniacally.
SHOT 08 — 5.6–6.4s | SNACK AISLE COLLISION
The lawnmower smashes into a towering cardboard display of potato chips.
The display collapses.
Hundreds of colorful snack bags fly toward the camera.
The operator ducks beneath falling merchandise.
SHOT 09 — 6.4–7.2s | DEVIL WOMAN REACTS
Crash zoom onto the devil woman.
She ducks behind the counter as bags rain down around her.
She looks straight into the camera with an expression of complete disbelief.
A potato chip bag slaps against the lens.
SHOT 10 — 7.2–8.0s | THE HEADLESS CUSTOMER
The automatic doors violently slide open again.
The Headless Horseman rides his enormous black horse directly into the convenience store, barely squeezing through the entrance.
His burgundy coat flaps.
He raises a glowing jack-o'-lantern.
The camera whips backward in panic.
SHOT 11 — 8.0–8.8s | HORSE FACE CLOSE-UP
The horse thrusts its enormous nostrils straight into the fisheye lens.
Extreme barrel distortion makes the snout look absurdly huge.
The horse snorts violently, fogging the lens.
The camera operator stumbles into a merchandise shelf.
SHOT 12 — 8.8–9.6s | SOMETHING IN THE COOLER
A strange metallic CLUNK comes from the refrigerated drink section.
The camera snaps toward the back wall.
One cooler door slowly swings open.
A tiny white shape is barely visible in the darkness near the floor.
For a split second, the chaos goes strangely quiet.
SHOT 13 — 9.6–10.4s | THE TINY GHOST
The tiny white-sheet ghost slowly floats out of the cooler.
It is barely knee-high.
Its two uneven black eyeholes face the camera.
The harsh flash makes its white sheet glow with ugly overexposure.
It tilts slightly to one side.
SHOT 14 — 10.4–11.2s | BOO
Smash cut to an extreme fisheye close-up of the tiny ghost.
It suddenly lunges a few inches toward the lens.
In a ridiculously small, squeaky voice, it whispers:
"BOO."
SHOT 15 — 11.2–12.0s | MONSTERS TERRIFIED
Violent whip-pan.
Dale the giant werewolf lets out a high-pitched terrified yelp and drops his dog treats.
All three skeletons recoil in panic, their jaws hanging open.
The devil woman's eyes go wide.
Everyone is genuinely horrified by this harmless little ghost.
SHOT 16 — 12.0–12.8s | TOTAL PANIC
Rapid handheld chaos.
The witch frantically reverses her lawnmower into a shelf.
The horse rears slightly and backs toward the automatic doors.
Dale scrambles away on all fours.
Skeletons collide with one another as they run.
The operator screams and tries to escape with them.
SHOT 17 — 12.8–13.4s | CAMERA DOWN
The orange-cap skeleton trips directly into the operator.
Its bones collapse into a chaotic pile.
The camera is knocked out of the operator's hands.
It tumbles toward the sticky convenience-store floor.
The picture spins violently, flash streaking across scattered bones and candy wrappers.
SHOT 18 — 13.4–14.0s | FINAL PAYOFF
The camera lies sideways on the floor.
The store behind it has gone almost completely black.
The tiny white ghost quietly floats into view.
It leans directly into the fisheye lens until its two crooked black eyeholes fill the entire frame.
One final tiny squeaky "Boo."
HARD CUT TO BLACK at exactly 14.0 seconds.
CAMERA BEHAVIOR — MANDATORY
The footage must feel like a single real camcorder recording, aggressively jump-cut into 18 frantic fragments.
The camera operator is physically present and reacts instinctively.
Camera movements should feel uncontrolled and human:
Extremely close fisheye framing.
Violent handheld shakes and whip-pans.
Sudden crash zooms.
Constant reframing as subjects move unpredictably.
People and creatures repeatedly colliding with or rushing toward the lens.
Accidental tilted horizons and imperfect framing.
Auto-exposure blowing out bright costumes and nearby faces.
Brief focus hunting after sudden movement.
Realistic motion smears during fast camera movement.
Abrupt hard cuts with no artificial transitions.
Every event must feel captured in the moment, never staged.
Maintain real physical collisions, object weight, believable impacts, messy inertia, and tangible environmental interaction.
The monsters should look physically present inside the store, not composited into the footage.
AUDIO — RAW CAMCORDER MICROPHONE ONLY
No music. No soundtrack. No cinematic score.
Only harsh, distorted, clipping early-2000s camcorder microphone audio.
Include:
Weak fluorescent electrical buzzing.
Automatic doors rattling and slamming.
Werewolf growls, panting, and a terrified yelp.
Claws scraping against the tile floor.
Dog treat bags ripping and crunching.
Skeleton bones clattering against the counter.
Soda bottle popping and liquid spraying.
People shouting "YO!" and screaming in panic.
A sputtering ride-on lawnmower engine.
Shelves collapsing and snack bags scattering.
Heavy horse hooves clopping on tile.
Loud horse snorting.
The refrigerator compressor humming.
The tiny ghost's unexpectedly squeaky "BOO."
The camera operator's panicked breathing.
The camera hitting the floor.
Explosive sounds should briefly overload the microphone.
Voices may clip, overlap, and distort naturally.
No voiceover or narration.
FINAL CREATIVE DIRECTION
The result should feel like someone found a completely unhinged Halloween camcorder tape recorded inside a cursed American convenience store in 2003.
Ugly, loud, frantic, overexposed, dirty, ridiculous, physically chaotic, and unintentionally hilarious.
The monsters must be genuinely intimidating in appearance but absurdly incompetent in behavior.
Never turn this into a polished cinematic horror trailer, a glossy commercial, a stylized music video, or a clean CGI animation.
The humor comes from the contrast between documentary-level realism and completely ridiculous supernatural behavior.
The final joke must land clearly: a terrifying collection of enormous Halloween monsters is completely defeated by one tiny, harmless-looking ghost.
14 seconds. 18 rapid shots. Extremely aggressive fisheye. Brutal direct flash. Chaotic found-footage realism. Hard cuts only. No music. No text.
@grok @imagine #Grok
🤖 Made with AI
One photo. Four fashion posters. 🤯
Same face. Same outfit. Different worlds.
I turned a street-style shot into a cohesive AI illustration series using ONE reusable prompt.
The trick: lock the character + style. Remix only the pose, prop & scene.
Prompt in the replies ↓
🤖 Made with AI
GPT Image 2 On ChatGPT
Prompt:
Create a finished editorial character poster inspired by the uploaded reference image. Preserve the recognizable identity, hairstyle, clothing details, pose, and important visual characteristics of the subject while translating them into a sophisticated flat geometric illustration.
FORMAT LOCK
Vertical 3:4 composition. Edge-to-edge warm off-white paper background. One standalone poster. Clean editorial layout.
CHARACTER
One adult character constructed from a few oversized asymmetric geometric shapes. Slightly exaggerated proportions, small head, elongated capsule-like limbs, simple mitten-style hands, and one oversized object integrated naturally into the pose and silhouette. Preserve the subject's recognizable appearance while simplifying details into graphic shapes.
STYLE
Flat geometric illustration, asymmetric proportions, crisp color blocks, soft airbrushed shading, subtle dark-to-color spray gradients where shapes overlap, fine print grain, delicate stippling mostly inside darker areas, refined contemporary editorial poster aesthetic.
FACE & EXPRESSION
Minimal black cut-paper facial features: simple crescent eyes and minimal mouth. The emotion should be communicated primarily through the character's full-body gesture and silhouette rather than detailed facial expression.
PALETTE
Warm off-white #F3F1EB, near-black 10100F, hot pink #F553D2, warm red #F42726, cobalt 293EA1, emerald 13785D. Keep colors flat and graphic with controlled soft gradients only where forms overlap.
Keep the character and oversized object as the main visual focus. Use generous negative space and a balanced editorial arrangement. The object should visibly affect the overall silhouette and feel integrated into the character's pose.
NEGATIVE PROMPT
No extra text, logos, outlined cartoon style, glossy 3D, realistic skin, photorealism, coarse canvas texture, busy background, extra limbs, distorted anatomy, excessive facial detail, photorealistic shading, random objects, clutter, watermark.
Prompt:
Create ONE premium editorial illustration poster based on the uploaded photograph.
Use the uploaded image as the primary visual reference. Transform its main subject into a sophisticated contemporary flat-geometric illustration while preserving its recognizable visual identity.
FORMAT
- Vertical 3:4 aspect ratio.
- ONE independent poster, never a collage or grid.
- Edge-to-edge warm off-white paper background.
- Premium contemporary editorial illustration aesthetic.
- Clean composition with generous negative space.
SUBJECT & IDENTITY
Identify the main person in the reference photograph.
Preserve their recognizable visual characteristics:
- Hairstyle, hair length, and hair color.
- Distinctive facial structure and appearance.
- Original outfit, clothing colors, and signature accessories.
- Overall body silhouette and personal style.
Translate these details into simplified graphic forms rather than realistic rendering.
Maintain visual identity through recognizable hair, clothing, accessories, and silhouette.
For a fashion poster series, maintain the same character design and outfit across every image.
CHARACTER DESIGN
Construct the character using bold, asymmetric geometric shapes.
Use:
- Slightly exaggerated editorial proportions.
- Elegant elongated limbs.
- Smooth, simplified anatomical forms.
- Flat shapes with subtle dimensional layering.
- Minimal facial details.
- Expressive body language.
- Strong, recognizable silhouettes.
The character should feel like an illustration from a premium independent fashion magazine.
ILLUSTRATION STYLE
Sophisticated 2D editorial illustration combining:
- Flat geometric color blocking.
- Crisp, clean shape boundaries.
- Asymmetric composition.
- Subtle airbrushed gradients.
- Delicate grain and stippling.
- Minimal cut-paper facial details.
- Refined retro-modern poster aesthetics.
Avoid overly realistic rendering or detailed anime-style illustration.
COLOR SYSTEM
Base palette:
- Warm ivory: #F3F1EB
- Near-black: #10100F
Select 2–3 complementary accent colors inspired by the uploaded image or the requested scene.
Optional accents:
- Hot pink: #F553D2
- Warm red: #F42726
- Cobalt blue: #293EA1
- Emerald green: #13785D
Keep colors bold, restrained, and visually harmonious.
COMPOSITION & HERO OBJECT
Integrate ONE visually dominant oversized object into the composition.
Choose an object that relates naturally to the uploaded photograph or requested scene, such as a handbag, umbrella, flower bouquet, coffee cup, camera, or another recognizable accessory.
The object should:
- Be noticeably exaggerated in scale.
- Interact naturally with the character's pose.
- Create a memorable graphic silhouette.
- Act as a strong visual focal point.
BACKGROUND
Use a clean, minimal graphic environment.
Suggest the setting with only a few simplified architectural or environmental shapes.
Avoid detailed photographic backgrounds.
CUSTOM SCENE VARIABLES
Scene: [AUTO / CUSTOM SCENE]
Character pose: [AUTO / CUSTOM POSE]
Oversized hero object: [AUTO / CUSTOM OBJECT]
Accent colors: [AUTO / CUSTOM COLORS]
When AUTO is selected, choose visually appropriate details based on the uploaded photograph.
SERIES CONSISTENCY
When producing additional posters:
- Preserve the same subject and recognizable visual identity.
- Maintain consistent illustration treatment.
- Reuse the same overall palette and paper texture.
- Keep consistent graphic proportions.
- Change only the requested pose, scene, and oversized object.
- Produce ONE independent poster per generation.
STRICT NEGATIVE PROMPT
No added text, typography, captions, logos, watermarks, photorealism, glossy 3D, anime rendering, comic outlines, extra limbs, duplicate characters, excessive facial detail, random props, cluttered backgrounds, or multi-panel layouts.
FINAL RESULT
A striking, collectible, contemporary flat-geometric editorial poster that feels professionally art-directed and visually consistent with a premium illustration series.
🤖 Made with AI
🍁🍂Turn any photo into a personal scrapbook page
— half real, half hand-drawn.
① Open GPT Image 2.5 / Nano Banana / Grok
② Upload your image
③ Paste the prompt
④ Generate your scene Prompt below 👇💬
🤖 Made with AI
GPT Image 2 On ChatGPT
PROMPT:
Create a vertical 3:4 editorial memory poster inspired by the uploaded photograph. Divide the composition into two horizontal sections: the top 50% preserves the original photograph with its recognizable subject, identity, pose, clothing, background, natural lighting, and realistic details. Do not redraw or stylize the top section.
The bottom 50% becomes a warm, hand-drawn illustrated scrapbook inspired by the photograph. Fill it with 7–9 small illustrations representing memorable details from the original scene, including the subject’s clothing, hairstyle, accessories, surrounding plants, flowers, trees, architecture, weather, activities, and other relevant objects. Arrange the illustrations in a balanced, playful collage with varied sizes and natural spacing.
ILLUSTRATION STYLE: Hand-drawn colored-pencil and watercolor sketches, delicate ink outlines, visible pencil strokes, imperfect artistic textures, soft coloring, charming doodles, tiny hearts, stars, arrows, and hand-drawn decorative accents. Use a warm cream-colored textured paper background with a nostalgic journal aesthetic.
COLOR PALETTE: Derive colors from the original photograph. Use soft, natural tones with subtle vintage warmth, earthy colors, and gentle highlights.
TYPOGRAPHY: Add a small handwritten title at the top of the illustrated section and a short, sentimental handwritten caption near the bottom. Keep the lettering elegant, lowercase, and naturally integrated into the artwork. Generate wording inspired by the photograph rather than using fixed text.
COMPOSITION RULES:
•Strict vertical 3:4 aspect ratio.
•Top and bottom sections occupy exactly 50% each.
•A perfectly straight horizontal division between sections.
•Original photograph remains photorealistic and unchanged.
•Bottom section contains illustrations related specifically to the uploaded image.
•Maintain a cohesive, handmade travel-journal or lifestyle-scrapbook aesthetic.
•High-quality textures, intricate sketch details, harmonious spacing, and professional editorial design.
IMPORTANT: Adapt every illustration, color, and handwritten caption to the uploaded photograph. Do not reuse objects or themes from reference examples unless they are relevant to the actual image. No extra photographs, no 3D elements, no glossy digital art, and no cluttered layout.
Result: A heartwarming, personalized photo-and-illustration poster that feels like a page from a beautifully illustrated personal diary.
PROMPT:
Create a vertical 3:4 editorial memory poster inspired by the uploaded photograph. Divide the composition into two horizontal sections: the top 50% preserves the original photograph with its recognizable subject, identity, pose, clothing, background, natural lighting, and realistic details. Do not redraw or stylize the top section.
The bottom 50% becomes a warm, hand-drawn illustrated scrapbook inspired by the photograph. Fill it with 7–9 small illustrations representing memorable details from the original scene, including the subject’s clothing, hairstyle, accessories, surrounding plants, flowers, trees, architecture, weather, activities, and other relevant objects. Arrange the illustrations in a balanced, playful collage with varied sizes and natural spacing.
ILLUSTRATION STYLE: Hand-drawn colored-pencil and watercolor sketches, delicate ink outlines, visible pencil strokes, imperfect artistic textures, soft coloring, charming doodles, tiny hearts, stars, arrows, and hand-drawn decorative accents. Use a warm cream-colored textured paper background with a nostalgic journal aesthetic.
COLOR PALETTE: Derive colors from the original photograph. Use soft, natural tones with subtle vintage warmth, earthy colors, and gentle highlights.
TYPOGRAPHY: Add a small handwritten title at the top of the illustrated section and a short, sentimental handwritten caption near the bottom. Keep the lettering elegant, lowercase, and naturally integrated into the artwork. Generate wording inspired by the photograph rather than using fixed text.
COMPOSITION RULES:
•Strict vertical 3:4 aspect ratio.
•Top and bottom sections occupy exactly 50% each.
•A perfectly straight horizontal division between sections.
•Original photograph remains photorealistic and unchanged.
•Bottom section contains illustrations related specifically to the uploaded image.
•Maintain a cohesive, handmade travel-journal or lifestyle-scrapbook aesthetic.
•High-quality textures, intricate sketch details, harmonious spacing, and professional editorial design.
IMPORTANT: Adapt every illustration, color, and handwritten caption to the uploaded photograph. Do not reuse objects or themes from reference examples unless they are relevant to the actual image. No extra photographs, no 3D elements, no glossy digital art, and no cluttered layout.
Result: A heartwarming, personalized photo-and-illustration poster that feels like a page from a beautifully illustrated personal diary.
Same prompt, two models — GPT vs Gemini
Left: GPT image 2.5
Right: Gemini
Which one do you prefer and why?
🤖 Made with AI
Same prompt, two models — GPT vs Grok
Prompt ⤵️
Create an ultra-realistic, high-end editorial street photography portrait of a stylish young woman standing in the middle of a busy urban pedestrian street, surrounded by a crowd of people. The image should look like a candid paparazzi photograph captured during a fashion week event, with cinematic lighting, natural movement, shallow depth of field, and premium fashion magazine aesthetics.
SUBJECT AND FACE:
A young adult woman with delicate facial features, a softly defined oval face, naturally fair skin with realistic skin texture, subtle rosy cheeks, a refined nose, naturally shaped eyebrows, and soft muted pink lips. She has a calm, confident expression with a slight closed-mouth smile. Her face is turned slightly toward the camera while maintaining direct eye contact through her sunglasses.
HAIR:
Long, voluminous dark brown hair with natural waves and a slightly tousled texture. Her hair falls freely over both shoulders and extends down her chest. Individual strands are visible, with a few loose flyaways catching the natural light. The hair should look realistic and naturally moving in the outdoor breeze.
SUNGLASSES:
Black rectangular designer-style sunglasses with glossy frames and dark tinted lenses. The glasses fit naturally on her face and reflect subtle environmental light. Preserve realistic frame proportions and reflections.
OUTFIT:
An oversized dark chocolate-brown wool coat with broad structured lapels, a relaxed luxury silhouette, thick premium fabric, and natural folds around the shoulders and sleeves. Underneath, she wears a simple black top visible near the neckline. The outfit is elegant, understated, sophisticated, and inspired by quiet luxury street fashion. No visible logos, text, or excessive accessories.
POSE AND BODY LANGUAGE:
A natural standing pose with her body angled slightly away from the camera while her face turns toward the lens. Her shoulders are relaxed, and her hands rest naturally near the front pockets of her coat. Her posture conveys quiet confidence and effortless sophistication. Frame her from approximately the waist up, with the subject positioned centrally in the composition.
BACKGROUND AND ENVIRONMENT:
A busy city sidewalk filled with pedestrians wearing dark winter coats and neutral-colored clothing. Include several people in the foreground and background, naturally out of focus. A partially visible person appears in the lower-left foreground, while another blurred person wearing sunglasses occupies the lower-right corner. Additional pedestrians fill the upper background. The crowd should create depth and a realistic urban atmosphere without distracting from the main subject.
LIGHTING AND COLOR GRADING:
Soft natural daylight with warm highlights falling on the woman's face and hair. Use subtle golden reflections, gentle shadows, rich chocolate-brown tones, deep blacks, and muted neutral colors. Maintain realistic skin tones and balanced exposure. Apply cinematic editorial color grading with a slightly warm, sophisticated appearance.
CAMERA AND COMPOSITION:
Professional full-frame mirrorless camera photography, 85mm portrait lens, aperture f/1.8, ISO 100–200, shutter speed 1/500 sec. Eye-level camera angle, precise focus on the subject's face, sunglasses, and hair, with strong optical background separation. Use a shallow depth of field and smooth natural bokeh. The main subject must be tack-sharp while foreground pedestrians and the distant crowd remain heavily blurred.
IMAGE QUALITY:
Photorealistic, high-resolution fashion photography, realistic skin pores, individual hair strands, detailed wool fabric, physically accurate shadows, natural facial proportions, authentic lens rendering, subtle film-like texture, premium luxury fashion campaign quality, candid street photography aesthetic.
ASPECT RATIO: Vertical 2:3 portrait composition.
NEGATIVE PROMPT:
Cartoon, anime, illustration, CGI appearance, plastic skin, excessive skin smoothing, artificial beauty filter, overexposed face, unnatural eyes, deformed hands, extra fingers, distorted sunglasses, asymmetrical facial features, unrealistic hair, oversaturated colors, sharp background, duplicated faces, artificial lighting, text, watermark, logos, low resolution, motion blur on the main subject.
Left: GPT image 2.5
Right: Grok
Which one do you prefer and why?
🤖 Made with AI
Ultra-Realistic Luxury Street Style Portrait
CHATGPT IMAGE
Prompt:
Create an ultra-realistic, high-end editorial street photography portrait of a stylish young woman standing in the middle of a busy urban pedestrian street, surrounded by a crowd of people. The image should look like a candid paparazzi photograph captured during a fashion week event, with cinematic lighting, natural movement, shallow depth of field, and premium fashion magazine aesthetics.
SUBJECT AND FACE:
A young adult woman with delicate facial features, a softly defined oval face, naturally fair skin with realistic skin texture, subtle rosy cheeks, a refined nose, naturally shaped eyebrows, and soft muted pink lips. She has a calm, confident expression with a slight closed-mouth smile. Her face is turned slightly toward the camera while maintaining direct eye contact through her sunglasses.
HAIR:
Long, voluminous dark brown hair with natural waves and a slightly tousled texture. Her hair falls freely over both shoulders and extends down her chest. Individual strands are visible, with a few loose flyaways catching the natural light. The hair should look realistic and naturally moving in the outdoor breeze.
SUNGLASSES:
Black rectangular designer-style sunglasses with glossy frames and dark tinted lenses. The glasses fit naturally on her face and reflect subtle environmental light. Preserve realistic frame proportions and reflections.
OUTFIT:
An oversized dark chocolate-brown wool coat with broad structured lapels, a relaxed luxury silhouette, thick premium fabric, and natural folds around the shoulders and sleeves. Underneath, she wears a simple black top visible near the neckline. The outfit is elegant, understated, sophisticated, and inspired by quiet luxury street fashion. No visible logos, text, or excessive accessories.
POSE AND BODY LANGUAGE:
A natural standing pose with her body angled slightly away from the camera while her face turns toward the lens. Her shoulders are relaxed, and her hands rest naturally near the front pockets of her coat. Her posture conveys quiet confidence and effortless sophistication. Frame her from approximately the waist up, with the subject positioned centrally in the composition.
BACKGROUND AND ENVIRONMENT:
A busy city sidewalk filled with pedestrians wearing dark winter coats and neutral-colored clothing. Include several people in the foreground and background, naturally out of focus. A partially visible person appears in the lower-left foreground, while another blurred person wearing sunglasses occupies the lower-right corner. Additional pedestrians fill the upper background. The crowd should create depth and a realistic urban atmosphere without distracting from the main subject.
LIGHTING AND COLOR GRADING:
Soft natural daylight with warm highlights falling on the woman's face and hair. Use subtle golden reflections, gentle shadows, rich chocolate-brown tones, deep blacks, and muted neutral colors. Maintain realistic skin tones and balanced exposure. Apply cinematic editorial color grading with a slightly warm, sophisticated appearance.
CAMERA AND COMPOSITION:
Professional full-frame mirrorless camera photography, 85mm portrait lens, aperture f/1.8, ISO 100–200, shutter speed 1/500 sec. Eye-level camera angle, precise focus on the subject's face, sunglasses, and hair, with strong optical background separation. Use a shallow depth of field and smooth natural bokeh. The main subject must be tack-sharp while foreground pedestrians and the distant crowd remain heavily blurred.
IMAGE QUALITY:
Photorealistic, high-resolution fashion photography, realistic skin pores, individual hair strands, detailed wool fabric, physically accurate shadows, natural facial proportions, authentic lens rendering, subtle film-like texture, premium luxury fashion campaign quality, candid street photography aesthetic.
ASPECT RATIO: Vertical 2:3 portrait composition.
NEGATIVE PROMPT:
Cartoon, anime, illustration, CGI appearance, plastic skin, excessive skin smoothing, artificial beauty filter, overexposed face, unnatural eyes, deformed hands, extra fingers, distorted sunglasses, asymmetrical facial features, unrealistic hair, oversaturated colors, sharp background, duplicated faces, artificial lighting, text, watermark, logos, low resolution, motion blur on the main subject.
Such a soft, dreamy aesthetic!
① Open GPT Image 2.5 / Nano Banana / Grok
② Upload your image
③ Paste the prompt
④ Generate your scene Prompt below 👇💬
🤖 Made with AI
GPT Image 2 On ChatGPT
PROMPT:
Convert the entire uploaded photograph into a clean, colorful, hand-drawn 2D cartoon illustration. Preserve the original composition, camera angle, poses, facial expressions, recognizable identities, hairstyles, clothing, accessories, and all important background elements.
ILLUSTRATION STYLE:
•Clean black outlines with slightly varied, hand-drawn line weight.
•Flat pastel colors with minimal cel shading.
•Simplified characters with cute proportions, expressive faces, and natural poses.
•Smooth, slightly imperfect lines for an authentic illustrated feel.
•Simplify realistic textures into clean shapes while retaining recognizable details.
•Recreate the original environment, including buildings, trees, fountains, pathways, and other objects, in the same cartoon style.
•Use a soft pastel palette with light blue skies, muted greens, warm skin tones, and gentle pinks.
•Add subtle artistic details while maintaining the original scene’s atmosphere.
PRESERVE THE ORIGINAL:
Keep the same number of people, their positions, body proportions, clothing colors, interactions, and relative sizes. Maintain the original background layout and perspective. Do not add, remove, or reposition important objects.
OUTPUT REQUIREMENTS:
A polished, full-frame 2D illustration with consistent outlines, flat colors, and a playful, storybook aesthetic. The entire image must be illustrated, with zero photorealistic elements.
AVOID: 3D rendering, photorealism, excessive shading, watercolor textures, blurry lines, distorted faces, extra fingers or limbs, changed outfits, altered poses, missing background details, and unwanted text or watermarks.
Aspect ratio: Match the uploaded photo exactly.
Pro tip: Upload one photo at a time and use this same prompt to transform different photos into a consistent cartoon style.
PROMPT:
Convert the entire uploaded photograph into a clean, colorful, hand-drawn 2D cartoon illustration. Preserve the original composition, camera angle, poses, facial expressions, recognizable identities, hairstyles, clothing, accessories, and all important background elements.
ILLUSTRATION STYLE:
•Clean black outlines with slightly varied, hand-drawn line weight.
•Flat pastel colors with minimal cel shading.
•Simplified characters with cute proportions, expressive faces, and natural poses.
•Smooth, slightly imperfect lines for an authentic illustrated feel.
•Simplify realistic textures into clean shapes while retaining recognizable details.
•Recreate the original environment, including buildings, trees, fountains, pathways, and other objects, in the same cartoon style.
•Use a soft pastel palette with light blue skies, muted greens, warm skin tones, and gentle pinks.
•Add subtle artistic details while maintaining the original scene’s atmosphere.
PRESERVE THE ORIGINAL:
Keep the same number of people, their positions, body proportions, clothing colors, interactions, and relative sizes. Maintain the original background layout and perspective. Do not add, remove, or reposition important objects.
OUTPUT REQUIREMENTS:
A polished, full-frame 2D illustration with consistent outlines, flat colors, and a playful, storybook aesthetic. The entire image must be illustrated, with zero photorealistic elements.
AVOID: 3D rendering, photorealism, excessive shading, watercolor textures, blurry lines, distorted faces, extra fingers or limbs, changed outfits, altered poses, missing background details, and unwanted text or watermarks.
Aspect ratio: Match the uploaded photo exactly.
Pro tip: Upload one photo at a time and use this same prompt to transform different photos into a consistent cartoon style.



































