Я все делал в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Gpt Image 2 и Seedance 2.0 (доступен от подписки PRO и выше)
Нужно: референс флага (ну либо просто текст).
Мой результат ищите в https://t.me/viviannru

FORMAT: 16:9 / 12s / 2 parts / hard cut / barbershop commercial / no dialogue. @Image1 is the first-part reference — the couple on the antenna spire with the black "ИДЕЛ BARBER" flag. @Image2 is the second-part reference — the wide GoPro view of the barber cutting hair on the observation platform high above the city. Match each reference exactly: same characters, same masks, same black outfits, identical flag design and logo. A single hard cut at 00:06, no morphing, no dissolve, no transition — part one ends and part two begins instantly.

[00:00-00:06] PART 1 — THE FLAG (helicopter telephoto, moving orbit). Scene first: exactly as @Image1 — the top of a tall steel antenna spire against a hazy blue sky. A man in a black balaclava, black cap and black outfit, and a woman in a black cat mask, both dressed in black, stand on the narrow spire structure, having just fastened the large black flag to the mast — the flag design references @Image1 exactly, cream "ИДЕЛ BARBER" logo clearly visible. The flag ripples and snaps in the high-altitude wind, fabric pulling taut then releasing. Camera: extreme long telephoto lens (600mm+) filmed from a helicopter that continuously orbits from left to right around the spire and also slowly cranes upward, gradually revealing the vast Manhattan cityscape far below through atmospheric haze — the spire, flag and the two people stay locked in the center of the frame the entire time. Heavy atmospheric compression, subtle heat-shimmer, image softly floating with the long-lens look, shallow depth of field. Around 00:02-00:04, a white television news helicopter (the type used by New York broadcast news networks) flies fast from left to right in the background behind the spire, passing quickly behind the mast and exiting the frame — it enters from the left edge, never spawns in the center. Audio: high wind, layered helicopter rotor thump swelling as the news chopper crosses, flag fabric snapping.

[00:06-00:12] PART 2 — THE HAIRCUT ABOVE THE CITY (static wide GoPro). Scene first: exactly as @Image2 — a locked static ultra-wide GoPro-style shot with mild fisheye distortion on the grated observation platform at the very top of the Empire State Building, the vast sunny Manhattan skyline and rivers sprawling far below on both sides, red aircraft beacon lights on the platform edge, the white steel antenna tower rising in the center. A black barber station table stands to the left with scissors, combs, clippers and grooming products. The man sits facing the camera in a large black barber cape, black face mask, white Crocs, only his dark hair and eyes visible. Behind him stands the woman in the black cat mask, black tank top and black cargo pants, working as the barber — she leans in and steadily trims and styles the hair on top of his head with clippers and comb, fluid confident motion, tiny hair clippings drifting away in the wind. Around 00:08-00:10, a black helicopter flies fast from left to right, passing very close behind the spire, seen half-side-on as it banks around and behind the platform, crossing the frame in about two seconds — but the woman does not react at all, calmly continuing the haircut. Camera: locked static wide GoPro shot, absolutely no camera movement, natural bright midday daylight. Audio: strong wind, soft snip of scissors and buzz of clippers, a fast rotor whoosh rising and fading as the black helicopter passes. Consistency: same man, same woman, same black cat mask, same black outfits, identical "ИДЕЛ BARBER" flag design between parts. Maintain stable facial and body proportions, natural fluid motion, no distortion.

P.S. Отправьте свои полученные кадры в chatgpt или Claude, в запросе отправьте и промпт, попросите адаптировать его под ваши кадры.

Я все делал в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Gpt Image 2 и Seedance 2.0 (доступен от подписки PRO и выше)
Нужно: фото человека, видео-референс танца.
Мой результат ищите в https://t.me/viviannru

@Image1 as the first frame and full scene reference — a city street sidewalk with pedestrians around, keep the environment, lighting, background and depth of field exactly as in @Image1. The person standing in the center of the frame references @Image2 for exact appearance: same face, same hairstyle, same clothing and accessories — keep identity fully consistent throughout. Scene first, then action: the person in the center starts dancing on the sidewalk, their choreography, movements and rhythm fully reference @Video1 — replicate the dance from @Video1 exactly, fluid, confident and on the beat. Background music and rhythm reference @Video1's audio. Around the dancer the world keeps living naturally: pedestrians walk past behind and cross in front of them in the foreground, going about their day, ignoring the dancer, while the person keeps dancing in the center of the frame. Camera: locked static medium-wide shot on a tripod, no movement, letting the dance carry the frame while pedestrians occasionally pass through the foreground. Cinematic quality, natural lighting matching @Image1, fine film grain. Constraints: maintain consistent facial features, hairstyle, clothing and accessories throughout, no warping or distortion, natural body proportions, smooth fluid motion, stable background, no flickering, no ghosting.

Я все делал в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Nano Banana 2 и Kling 3.0
Нужно: фото машины
Мой результат ищите в https://t.me/viviannru

*Важно! Промпты универсальные, подходят для любой машины
*Смотрите скрины, чтобы понимать, куда и какие изображения я отправлял.

Я все делал в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Gpt Image 2 и Seedance 2.0 (доступен от подписки PRO и выше)
Нужно: фото машины, фото водителя, фото пилота.
Мой результат ищите в https://t.me/viviannru

*Важно! Промпты универсальные, подходят для любой машины и людей (кроме суперзвезд)
*Смотрите скрины, чтобы понимать, куда и какие изображения я отправлял.

【Style】Ultra-cinematic blockbuster cinematography in the tradition of Michael Bay action films, shot on professional cinema camera with anamorphic lenses, 2.39:1 widescreen, shallow depth of field, dynamic motion blur, deep rich color grade, sharp highlights, golden bright daylight, lens flares on the sun, high dynamic range, professional film quality. 【Duration】15 seconds 【Native Audio】Powerful cinematic engine roar building in pitch, screaming jet afterburner howl, rushing wind, tire roar, deep bass impact, no dialogue. [00:00-00:03] Shot 1: THE DRIVER (Side Close-up, through side window). Bright daytime. Side profile of the car driver (face and outfit reference @Image2) seated inside the car (interior matches the car from @Image1), shot through the side window from a low close angle. The driver grips the wheel, jaw tight, eyes locked forward with fierce determination, body subtly vibrating from extreme speed. Sun glints sharply across the glass. Tight slow push-in. Engine roar builds. [00:03-00:06] Shot 2: THE PILOT (Side Close-up, matching angle). Cut to the same framing logic: side profile of a fighter-jet pilot (face clearly reference @Image3) inside the cockpit of a Russian Su-57 fighter jet. The pilot's helmet visor is fully raised, his entire face clearly visible and well-lit, eyes intense and focused, sweat on his brow. Bright daylight fills the cockpit, sun reflecting on the canopy. Tight slow push-in. Jet howl rises in the mix. [00:06-00:08] Shot 3: BACK TO THE DRIVER (Side Close-up). Cut back to the driver (@Image2) through the side window, same tight side profile, his expression hardening, knuckles white on the wheel, intensity peaking. Slow push-in. Engine and jet noise both layered loud. [00:08-00:13] Shot 4: THE RACE (Crane Pull-Back to Front-Side Wide). Smooth cinematic crane move: the camera lifts and glides backward away from the driver, pulling out to a dynamic front three-quarter angle revealing the full chase on a long straight runway in bright daylight. On the RIGHT, the car from @Image1 hurtles forward at insane speed, wheels a blur, asphalt streaking, faint tire smoke. On the LEFT, a Russian Su-57 fighter jet FLIES low through the air just above the car's height, afterburners blazing, not touching the ground, ripping through the sky beside it. Extreme speed and energy: roadside markers, grass, debris and dust streak past and fling violently backward, heavy motion blur on everything except the two subjects. Camera locks to a front three-quarter tracking position, both racing toward and past the lens at blistering velocity, neck and neck. Overwhelming engine and jet sound. [00:13-00:15] Shot 5: THE WIN (Same Front-Side Angle, fixed). Camera holds the exact same front three-quarter level. The car (@Image1) slams the throttle and surges forward with a burst of raw power, pulling decisively ahead of the Su-57, the jet visibly falling behind into the background as the car dominates the frame. Maximum dynamic blur, dust exploding behind it. Engine roar peaks. Triumphant final daytime frame. Car appearance, color and details stay identical to @Image1 throughout. Driver identity matches @Image2, pilot face clearly matches @Image3 with raised visor and visible face, faces stable with no distortion. Su-57 design consistent and always airborne. Cinematic professional camera look maintained the whole time.

Я все делаю в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Seedance 2.0 (доступен от подписки PRO и выше)
Нужно: фото ребенка при желании
Мой результат ищите в https://t.me/viviannru

Два вариант, можно просто отправить промпт и тогда нейросеть сама создаст ребёнка, либо можно добавить фото своего ребенка.

@Image1 is the baby character reference. The baby's face, skin tone, hair, and patterned onesie are identical to @Image1 throughout the entire video, consistent in every frame, no morphing, no distortion. 【Style】Authentic smartphone home video, documentary realism, shot handheld on a modern phone in 9:16 vertical. Soft natural daylight from off-frame windows, warm cozy living room, cream shaggy rug, neutral grey sofa and toys blurred in the background. Natural phone-camera look: mild handheld sway, slight breathing motion, realistic depth of field, true-to-life colors, no cinematic grade, no film grain, 100% real-life smartphone texture. 【Duration】14 seconds 【Native Audio】Quiet home room tone throughout: faint ambient hum, soft rug rustle, the baby's little breaths and coos, a small single puppy bark, gentle wet licking sounds, a soft baby giggle near the end. A mother's warm off-screen voice softly says in English at 0:11 "Aww, he loves you!". No music. One continuous unbroken take, no cuts, filmed entirely from the mother's point of view holding the phone. The baby (@Image1) sits on the soft cream rug at home, a small fluffy golden retriever puppy with beautiful bright blue eyes beside him. 0:00-0:03: The puppy steps in close, sniffing toward the baby. The baby leans his head down curiously and his forehead gently bumps the top of the puppy's head — a soft light bump. 0:03-0:05: Startled by the little bump, the puppy lets out one small soft bark and quickly backs away a short distance, ears slightly back, its blue eyes wide. The baby's face shows pure confusion, eyebrows raised, mouth slightly open, not understanding what just happened. 0:05-0:07: The puppy watches the baby from a little distance, head tilted, clear blue eyes fixed on the baby. The baby stares back at the puppy, still puzzled, blinking. 0:07-0:10: The puppy quickly trots back to the baby and starts sweetly licking the baby's face, like little kisses. The baby freezes in surprise, eyes wide. 0:10-0:14: The baby's expression melts from confusion into delight — wide eyes, a big surprised smile, then a soft happy giggle as he looks at the puppy. At the same moment he reaches up and clumsily, in a babyish way, pats the puppy on top of its head. The puppy looks happy and content, its blue eyes bright, leaning slightly into the pat. Baby facial features and onesie stay identical to @Image1 in every shot, natural baby skin texture, the puppy keeps the same beautiful blue eyes consistently in every frame, realistic puppy fur physics reacting to airflow and the baby's pats, all movements soft, gentle and physically natural, no stiffness, no warping. Handheld smartphone framing follows the baby and puppy naturally with small realistic adjustments, never a professional smooth gimbal move. Authentic candid family-phone-video feel.

Я все делаю в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Seedance 2.0 (доступен от подписки PRO и выше)
Нужно: -
Мой результат ищите в https://t.me/viviannru

FORMAT: 15s, 4:3 aspect ratio, single locked static camera (smartphone handheld footage, slight natural micro-shake, vertical-ish phone framing feel, no camera movement whatsoever), heavily sped-up timelapse pacing, absurd comedy montage, no dialogue. STYLE: amateur smartphone video look, bright sunny afternoon daylight, long hard shadows stretching diagonally across the asphalt, slightly overexposed sky, real-life photographic texture (100% real shooting, suppress CGI tells), motion blur on fast objects, subtle phone-camera grain and slight compression, photoreal. SCENE (built first, stays fixed the entire video): a wide corner apex of a professional racing circuit shot from outside the track behind a tall metal safety fence, grassy slope rising behind the fence, grey asphalt with faded racing lines and tire marks, the camera is locked on a tripod-like static frame pointed at the apex, NEVER moves, every object enters the corner from frame-left, slides through the apex in an aggressive controlled drift, and exits frame-right, each leaving smoke trails, sparks or dust, all motion extremely fast due to timelapse playback. [00:00-00:01] Shot 1: THE FAKE-OUT. A blue Bugatti hypercar enters the corner in a fast drift, rear tires erupting in white smoke, then a dark Formula 1 car snaps through the apex right behind it tires smoking — looks like a normal race broadcast. [00:01-00:02] Shot 2: THE BREAK. The massive grey stern of a giant cargo container ship slides sideways through the corner trailing thick smoke, dwarfing everything, then a red double-decker London bus drifts behind it. [00:02-00:03] Shot 3: THE CASTLE. An entire stone medieval castle with towers drifts through the apex kicking up dust, followed by a red London telephone booth sliding fast. [00:03-00:04] Shot 4: THE WEIRD. A giant smooth mirrored chrome sphere rolls and drifts through, reflecting the track, then a man calmly sitting inside a white bathtub slides through the corner, a low-flying helicopter skimming just above him. [00:04-00:05] Shot 5: THE FLOAT. A futuristic space shuttle capsule drifts through trailing blue flame, beside it a giant yellow rubber duck slides sideways, and a flying purple grand piano skids through the apex. [00:05-00:06] Shot 6: THE WHALE. A gigantic blue whale slides on its belly across the asphalt at insane speed, sparks flying, followed by a fully decorated Christmas tree drifting through with tinsel flying off. [00:06-00:07] Shot 7: THE RACE REMINDER. A Formula 1 car flashes through the corner normally tires smoking, immediately followed by a giant flaming hamburger and a drifting yellow school bus. [00:07-00:08] Shot 8: THE CRANE. A white GT race car drifts through, beside it a tall construction crane on tracks dragging a green construction container on cables, dust everywhere, and a runaway shopping trolley sliding through. [00:08-00:09] Shot 9: THE TINY ONES. A baby stroller races through the apex at full speed, and a wooden park bench drifts past shooting blue jet flames from underneath, followed by an old red telephone booth again. [00:09-00:10] Shot 10: THE MUSEUM. A massive golden-framed oil painting slides sideways through the corner, a white LMP prototype race car drifts right behind it, and a tall wooden grandfather clock skids through swinging its pendulum. [00:10-00:11] Shot 11: THE ZOO. A drifting giant pink flamingo statue slides through, a porta-potty toilet cabin skids sideways with smoke, and a flying flying-saucer UFO hovers through the apex. [00:11-00:12] Shot 12: THE KITCHEN. A giant floating bathtub-sized cup of coffee with steam drifts through, an inflatable bouncy castle slides sideways, and a vending machine skids across the asphalt sparking. [00:12-00:13] Shot 13: THE SWARM. Three race cars drift through almost touching in tight close racing combat, tires smoking heavily, fighting for position. [00:13-00:14] Shot 14: THE OCTOPUS. A gigantic red octopus slides across the track on its tentacles trailing slime, then a galloping cartoonish dinosaur (T-rex) drifts through the corner. [00:14-00:15] Shot 15: THE FINALE. A final burst of fast race cars drifts through the apex one after another tires smoking, then the track empties as a lone red London telephone booth slides through last and exits frame-right. The camera stays absolutely locked and motionless the entire time. Realistic drift physics: tire smoke disperses naturally, sparks bounce with correct physics, dust plumes rise realistically, hard sunlight casts long consistent shadows for every object on the asphalt, motion blur on all fast-moving objects. Sped-up timelapse pacing throughout, continuous nonstop flow of objects, each entering left and exiting right through the same drift line. No camera movement, no zoom, no pan. Smartphone footage realism, sunny day, photoreal.

Я все делаю в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Nano banana 2 и Seedance 2.0 (доступен от подписки PRO и выше) Нужно: фото машины.
Мой результат ищите в https://t.me/viviannru

Я все делал в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Gpt Image 2 и Seedance 2.0 (доступен от подписки PRO и выше) Нужно: фото вашего ребенка. (Но подойдёт и для взрослых)
Мой результат ищите в https://t.me/viviannru

Возьми внешность человека из фото 1 фото 2 и Create a professional photorealistic character reference sheet (character turnaround sheet) of the SAME person from the uploaded reference photo(s) 1-2. If multiple reference images are provided, carefully analyze ALL of them together to build a complete and accurate understanding of the person's appearance, face, hair, clothing and accessories from different angles, and use this combined information to generate the turnaround consistently.

CRITICAL — preserve identity 100%:

Keep the face absolutely identical to the reference: same facial features, same face shape, same eye shape and color, same nose, same lips, same eyebrows, same skin tone, same age, same makeup (if any).
Keep the hair 100% identical: same color, same length, same hairstyle, same parting, same texture, same volume.
Do NOT beautify, do NOT change proportions, do NOT make the person younger or older, do NOT alter ethnicity. The person must be perfectly recognizable in every shot.
CRITICAL — preserve outfit and accessories EXACTLY as in the reference:

Keep the clothing 100% identical to the reference photo(s): same garment, same color, same cut, same fabric, same details, same prints, same stripes, same logos, same brooches/pins, same neckline. Do not add a jacket, blazer, scarf, hat, sunglasses or any other clothing item that is not visible in the reference.
Keep all accessories 100% identical: same earrings, same necklace, same rings, same watch, same brooch — exactly as shown in the reference.
Do NOT add any new accessories. Do NOT remove existing accessories.
If only the upper body is visible in the reference, logically extend the outfit downward in a neutral, realistic way that matches the style and color palette of the visible clothing (for example simple matching trousers/skirt and plain shoes), keeping it minimal and non-distracting.
Style: hyper-realistic photography, studio quality, 8K, sharp focus, professional color grading, cinematic soft lighting, shot on Hasselblad, 85mm lens, clean neutral light grey studio background, soft even lighting without harsh shadows. Consistent lighting and background across all 6 shots.

COMPOSITION — single reference sheet, overall canvas aspect ratio 16:9 (horizontal), divided into a strict, clean grid with thin neutral dividers between panels:

LEFT HALF of the canvas (occupies 50% of the width, full height) — split vertically into 2 EQUAL TALL VERTICAL PANELS (full-body shots, head to toes, character fills the panel from top to bottom with small headroom):

Left vertical panel — Full body, FRONT view: standing straight, natural relaxed confident pose, arms relaxed at sides, looking at camera.
Right vertical panel — Full body, BACK view: exact same standing pose seen from directly behind, same scale and same vertical alignment as the front view, showing the back of the outfit and hair.
RIGHT HALF of the canvas (occupies the other 50% of the width, full height) — split into a 2x2 GRID of 4 EQUAL PANELS (close-up portraits, head and shoulders, tight framing on the face):
3) Top-left panel — Face CLOSE-UP, FRONT view: head and shoulders, looking straight at camera, neutral calm expression.
4) Top-right panel — Head CLOSE-UP, BACK view: directly from behind, showing the back of the head and hairstyle in full detail.
5) Bottom-left panel — Face CLOSE-UP, THREE-QUARTER view from the RIGHT side (face turned slightly so we see more of the right side of the face).
6) Bottom-right panel — Face CLOSE-UP, THREE-QUARTER view from the LEFT side (face turned slightly so we see more of the left side of the face).

All 6 shots must show the SAME person, SAME outfit, SAME accessories, SAME hairstyle, SAME lighting, SAME background — only the camera angle and framing changes. Consistent character design across all frames, like an official movie character turnaround or a fashion lookbook.

Layout rules: keep the 16:9 master frame; left side = 2 vertical full-body panels of equal width and full height; right side = 2x2 grid of equal close-up panels; thin clean separators between all panels; no text, no labels, no watermarks, no logos, no extra characters, no duplicates of the person inside a single panel.

Ultra-detailed, photorealistic skin texture, realistic fabric texture, realistic hair strands, natural shadows. The final result must look like six real photographs of the same real person taken in one studio session, assembled into one professional 16:9 character reference sheet.

@Image1 is the character reference for the person in the video. The person has identical facial features, hairstyle, build, and skin tone to @Image1 across every second of the clip. Same face throughout. Their clothing is EXACTLY the clothing visible in @Image1 — do not change, replace, or modify any garment. 【Style】Authentic vertical smartphone video, 9:16 aspect ratio, 1080x1920, shot handheld on a modern smartphone at night. Documentary-style mobile footage aesthetic — natural handheld camera shake, slight motion blur during movement, realistic smartphone low-light noise and grain. Maximum photorealism — looks like a real video filmed by a friend, NOT a movie, NOT cinematic, NOT stylized. Atmospheric night mood with natural lighting from street lamps only. 【Duration】10 seconds 【Native Audio】Quiet urban park ambience at night — distant city traffic, faint wind through bare branches. Soft crackling of burning birthday candles. At the climax: a sharp solid "thunk" as the glass cola bottle's bottom strikes the wooden bench, immediately followed by a loud realistic POP of the cola cap shooting off and the hissing fizz of cola fountaining, simultaneously combined with a deep "WHOOOMP" fireball roar as a huge flame erupts. Soft laughter from the person from @Image1 at the end. NO dialogue, NO speech. NO music, NO soundtrack — only natural diegetic sounds. 【Setting】Nighttime urban city park. Background: dark sky, bare leafless trees, distant warm street lamps creating soft natural bokeh, dry ground with scattered dirt and dead leaves. An old weathered wooden park bench sits in the center of the frame. Realistic, ordinary park — nothing magical, nothing fantastical. 【Camera】Handheld smartphone shot, vertical orientation. The camera is held by an unseen friend filming. Pronounced organic handheld shake throughout. Camera framing roughly waist-to-head level. NOT stabilized, NOT gimbal-smooth. SEQUENCE OF ACTIONS (strict timing, 10 seconds total): [00:00-00:01.5] The shot opens on an empty weathered wooden park bench centered in the frame, nighttime park setting. The person from @Image1 enters from the right side of the frame, walking briskly toward the bench, smiling warmly at the camera but NOT speaking. In their RIGHT hand they hold a clear glass bottle of cola — classic dark brown carbonated cola visible through the transparent glass, sealed with a metal cap on top — held vertically (cap pointing up, base pointing down). In their LEFT hand they carry a small round LIGHT-COLORED birthday cake (pale cream / white frosting) on a flat plate, with a single tall number-shaped candle in the form of the digit "AGE_NUMBER" standing upright on top, burning with a strong bright orange flame, flickering noticeably as they walk. Clothing matches @Image1 exactly. [00:01.5-00:03] They reach the bench and sit down in the center of frame, facing the camera directly, smiling confidently, silent. They place the light cream cake with the burning number "AGE_NUMBER" candle on their left palm extended out to their LEFT side (RIGHT side of the frame from camera POV), holding the glass cola bottle in their right hand on the opposite side. The number-shaped candle flame burns brightly and tall throughout. [00:03-00:04] The person from @Image1 shakes the sealed glass cola bottle vigorously with their right hand — cola visibly fizzing and foaming inside the transparent glass. They glance at the camera with a playful smile. The cake stays steady in their left hand, the number "AGE_NUMBER" candle still burning brightly. [00:04-00:05] CRITICAL SIMULTANEOUS ACTION — read carefully: The person holds the sealed glass cola bottle in their RIGHT hand in a fully VERTICAL position — cap pointing straight UP, base pointing straight DOWN. They raise their right arm and forcefully SWING IT STRAIGHT DOWNWARD in a hammering vertical motion, keeping the bottle VERTICAL throughout the swing. The flat BOTTOM (BASE) of the glass cola bottle slams DOWNWARD onto the wooden bench seat on their right side (LEFT side of the frame from camera POV). The bottle does NOT swing sideways, does NOT rotate — purely vertical hammering strike. A sharp "thunk" on impact. AT THE EXACT SAME INSTANT as the bottle impact, the person sharply TURNS THEIR HEAD to their LEFT (toward the cake on the right side of the frame), takes a deep breath, and forcefully BLOWS OUT a thick cloud of fine flammable powder (like a fire performer / fire breather using lycopodium or powdered sugar) from their mouth, AIMED directly across the top of the burning number candle. The powder cloud instantly ignites from the candle flame, exploding into a MASSIVE bright orange-yellow fireball that bursts horizontally outward to the right side of the frame, roughly 1 to 1.5 meters wide, exactly like a circus fire breather's flame trick. [00:05-00:08.5] EXTREME SLOW-MOTION (approximately 1/8 speed). TWO simultaneous spectacular effects unfold in slow motion: (1) From the glass cola bottle on the LEFT side of the frame — the metal cap shoots vertically off the bottle top, and a powerful cola fountain erupts straight UP, approximately 1.5 to 2 meters tall, a tall column of dark caramel-brown foam and individual liquid droplets, realistic carbonated liquid physics, droplets reflecting ambient street lamp light naturally. (2) From the person's mouth on the RIGHT side of the frame — a HUGE bright orange-yellow fireball blooms outward horizontally above the light cream cake, with visible billowing flame structure, dark smoke trailing, hot embers and tiny glowing particles inside the fire (only natural combustion sparks, not magical), the fireball expanding and curling in real fire physics. The light cake stays in their left hand with the number "AGE_NUMBER" candle still burning visibly through the flame. The person from @Image1 keeps the bottle vertical in their right hand and the cake extended in their left hand, head turned to their left toward the cake, mouth open mid-blow, eyes squinted from the heat. Slow motion captures every cola droplet and every flame tongue moving slowly through the cool night air. NO glowing magical particles, NO sparkles, NO fireworks, NO confetti — only realistic carbonated cola and realistic combustion fire. [00:08.5-00:10] Slow-motion ends — footage returns to normal real-time speed. The fireball quickly dissipates into thin smoke rising upward; the number "AGE_NUMBER" candle on the light cake is still burning (or partially extinguished by the blast, smoking lightly). Cola foam and droplets rapidly fall downward, some landing on the person's shoulders, hair, and clothing, some hitting the bench. The fountain weakens but keeps fizzing. The person sits relaxed on the bench, glass cola bottle still in their right hand, light cream cake still in their left hand, laughing softly and smiling widely at the camera, cola dripping off them, faint smoke curling around their face. Final beat holds until 00:10. ABSOLUTE CONSTRAINTS — READ CAREFULLY: 1. CHARACTER (@Image1): Face, hair, build, and skin of the person from @Image1 stay identical to @Image1 every frame. Clothing is EXACTLY as in @Image1 — no changes, no extra accessories. 2. NO SPEECH: The person never speaks, only smiles and softly laughs. 3. PROPS IN HANDS: RIGHT hand holds the clear glass cola bottle the entire clip. LEFT hand holds a small round LIGHT-COLORED (cream/white frosting) birthday cake with a single tall number-shaped candle in the form of the digit "AGE_NUMBER" burning on top the entire clip. Both objects stay in hand throughout — they never drop the cake, never drop the bottle. 4. CANDLE: The number "AGE_NUMBER" candle burns brightly with a tall flickering orange flame from 00:00 until the fireball at 00:04. After the fireball it may be blown out and smoking, but the cake itself stays intact. 5. BOTTLE STRIKE — CRITICAL MOTION: The glass cola bottle is held VERTICALLY (cap up, base down); right hand grips the bottle around the middle/upper portion; arm swings STRAIGHT DOWNWARD in a vertical hammering motion; bottle stays VERTICAL during the entire swing — does NOT tilt sideways, does NOT rotate; flat BOTTOM (BASE) slams DOWN onto the wooden bench seat; strike location: bench seat on the person's RIGHT side (LEFT side of frame from camera POV). 6. FIRE BREATHING — CRITICAL MOTION: At the EXACT same instant as the bottle impact, the person turns their head to the LEFT (toward the cake on the right side of frame), and blows a cloud of flammable powder from their mouth over the burning number candle. The powder ignites instantly into a large horizontal fireball about 1–1.5 meters wide, just like a real circus fire breather / fire performer. The fire blooms AWAY from their face (to the right side of the frame), never burning the person. The cake remains in their left hand undamaged. 7. SIMULTANEOUS TIMING: The bottle impact (cola eruption) and the mouth fire breath (fireball) happen at the SAME instant — frame 00:04. Both effects then play out together in slow motion. 8. FOUNTAIN HEIGHT: Cola fountain approximately 1.5 to 2 meters tall, a HIGH powerful vertical column of dark caramel-brown carbonated foam. Realistic. 9. FIREBALL: Realistic orange-yellow combustion flame with billowing structure, dark smoke, real fire physics. NO blue magical fire, NO cartoon flames, NO sparkles. Just real fire-breather flame. 10. NO MAGICAL EFFECTS — STRICT REALISM: NO glowing magical particles, NO sparkles, NO fireworks, NO bokeh light particles in the foreground, NO luminescent fantasy effects, NO confetti. ONLY realistic liquid cola and realistic combustion fire. Maximum photorealism, like real smartphone footage. 11. SLOW-MOTION: Begins at 00:05 (immediately after the combined bottle impact + fireball ignition), runs at ~1/8 speed, ends at 00:08.5. Final 1.5 seconds (00:08.5–00:10) returns to real-time speed. 12. AESTHETIC: Authentic handheld smartphone footage. NOT cinematic, NOT movie-quality, NOT stylized. Raw realistic phone video. 13. AUDIO: Only natural diegetic sounds — bench thunk, cap pop, cola hiss, fireball "whoomp" roar, soft laughter, candle crackle, park ambience. NO music, NO score, NO speech. 14. NO TEXT OVERLAYS, NO SUBTITLES, NO WATERMARKS anywhere in the frame.

Я все делал в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Seedance 2.0 (доступен от подписки PRO и выше)
Нужно: фото упаковки и еды.
Мой результат ищите в https://t.me/viviannru

*Я делал в максимально простом качестве, использовал самый дешевый вариант (31 токенов), чтобы было похоже на оригинал. При необходимости, можете настройки подогнать по себя.
*Есть два варианта промпта, с референсным фото и без.

@Image1 is the product reference. Read the packaging design, brand logo, and exact type of food shown in @Image1, and keep that same food and that same branded package identical throughout the video. FORMAT: 10s vertical 9:16, single continuous handheld take, no cuts. STYLE: amateur smartphone home video, slightly soft focus, mild digital noise and compression, warm indoor tungsten lighting from above, slightly underexposed, casual spontaneous feel, not cinematic. A Slavic baby girl, around one year old, fair skin, light blonde wispy hair, big grey-blue eyes, wearing a soft pink long-sleeve shirt, sits in a black leather chair at a home dining table with beige wooden-panel walls behind her, facing the camera in a tight medium close-up. [00:00-00:02] The open branded package from @Image1 is visible in the lower part of the frame; an adult's hand reaches into it, takes out one piece of the food from @Image1, and holds it up toward the baby. [00:02-00:03] The hand offers the food to the baby's mouth; she grins with excited anticipation, leans forward, opens her mouth wide and takes a big eager bite, cheeks puffing as she pulls back. [00:03-00:05] She closes her eyes tightly and scrunches her whole face, eyebrows pulling together, lips pursing as she chews slowly with visible confusion, jaw working side to side, unsure of the taste; the package drifts out of frame. [00:05-00:06] Still chewing, her eyes slowly begin to open, her scrunched expression gradually softens, eyebrows relaxing as understanding starts to form. [00:06-00:08] Her eyes widen fully, she turns her head slightly toward the person off-frame right who gave her the food, staring in stunned realization that it tastes amazing. [00:08-00:10] Her eyebrows lift high, her mouth drops open in an exaggerated gasp of delighted surprise, eyes huge and round, frozen in pure astonishment. Camera: aggressive handheld shake throughout as if someone suddenly grabbed a phone and started filming on impulse — at 00:00 the frame jolts and whips into position with motion blur, struggling to find the package and the baby, then keeps shaking with frequent sharp jerks, quick reframing wobbles and small drifts that never fully settle. Audio: quiet home room ambience, soft chewing sounds, faint crinkle of the package, faint rustle of the person holding the phone, no music, no dialogue. Natural toddler skin texture, consistent face and expressions matching each emotional beat, consistent lighting and background, the food and packaging exactly matching @Image1, no distortion.

FORMAT: 10s vertical 9:16, single continuous handheld take, no cuts. STYLE: amateur smartphone home video, slightly soft focus, mild digital noise and compression, warm indoor tungsten lighting from above, slightly underexposed, casual spontaneous feel, not cinematic. A Slavic baby girl, around one year old, fair skin, light blonde wispy hair, big grey-blue eyes, wearing a soft pink long-sleeve shirt with a small bunny print, sits in a black leather chair at a home dining table with beige wooden-panel walls behind her, facing the camera in a tight medium close-up. An adult's hand enters from the lower-right holding a triangular slice of cheese pizza, golden crust with browned cheese bubbles. [00:00-00:02] The baby grins with excited anticipation, leans forward, opens her mouth wide and takes a big eager bite of the pizza tip, cheeks puffing as she pulls back. [00:02-00:04] She closes her eyes tightly and scrunches her whole face, eyebrows pulling together, lips pursing as she chews slowly with visible confusion, jaw working side to side, unsure of the taste. [00:04-00:06] Still chewing, her eyes slowly begin to open, her scrunched expression gradually softens, eyebrows relaxing as understanding starts to form. [00:06-00:08] Her eyes widen fully, she turns her head slightly toward the person off-frame right who gave her the pizza, staring in stunned realization that it tastes amazing. [00:08-00:10] Her eyebrows lift high, her mouth drops open in an exaggerated gasp of delighted surprise, eyes huge and round, a thin string of cheese hanging from her lower lip, frozen in pure astonishment. Camera: aggressive handheld shake throughout as if someone suddenly grabbed a phone and started filming on impulse — at 00:00 the frame jolts and whips into position with motion blur, struggling to find the baby, then keeps shaking with frequent sharp jerks, quick reframing wobbles and small drifts that never fully settle. Audio: quiet home room ambience, soft chewing sounds, faint rustle of the person holding the phone, no music, no dialogue. Natural toddler skin texture, consistent face and expressions matching each emotional beat, consistent lighting and background, no distortion.

Я все делал в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Seedance 2.0 (доступен от подписки PRO и выше)
Нужно: фото пакета.
Мой результат ищите в https://t.me/viviannru

*В статье найдет 3 варианта промпта:
1. Можете отправить фото своей девушки
2. Девушка в черном комбинезоне
3. Девушка в шортах

Возьми внешность девушки из фото 1 фото 2 и Create a professional photorealistic character reference sheet (character turnaround sheet) of the SAME person from the uploaded reference photo(s) 1-2. If multiple reference images are provided, carefully analyze ALL of them together to build a complete and accurate understanding of the person's appearance, face, hair, clothing and accessories from different angles, and use this combined information to generate the turnaround consistently.

CRITICAL — preserve identity 100%:

Keep the face absolutely identical to the reference: same facial features, same face shape, same eye shape and color, same nose, same lips, same eyebrows, same skin tone, same age, same makeup (if any).
Keep the hair 100% identical: same color, same length, same hairstyle, same parting, same texture, same volume.
Do NOT beautify, do NOT change proportions, do NOT make the person younger or older, do NOT alter ethnicity. The person must be perfectly recognizable in every shot.
CRITICAL — preserve outfit and accessories EXACTLY as in the reference:

Keep the clothing 100% identical to the reference photo(s): same garment, same color, same cut, same fabric, same details, same prints, same stripes, same logos, same brooches/pins, same neckline. Do not add a jacket, blazer, scarf, hat, sunglasses or any other clothing item that is not visible in the reference.
Keep all accessories 100% identical: same earrings, same necklace, same rings, same watch, same brooch — exactly as shown in the reference.
Do NOT add any new accessories. Do NOT remove existing accessories.
If only the upper body is visible in the reference, logically extend the outfit downward in a neutral, realistic way that matches the style and color palette of the visible clothing (for example simple matching trousers/skirt and plain shoes), keeping it minimal and non-distracting.
Style: hyper-realistic photography, studio quality, 8K, sharp focus, professional color grading, cinematic soft lighting, shot on Hasselblad, 85mm lens, clean neutral light grey studio background, soft even lighting without harsh shadows. Consistent lighting and background across all 6 shots.

COMPOSITION — single reference sheet, overall canvas aspect ratio 16:9 (horizontal), divided into a strict, clean grid with thin neutral dividers between panels:

LEFT HALF of the canvas (occupies 50% of the width, full height) — split vertically into 2 EQUAL TALL VERTICAL PANELS (full-body shots, head to toes, character fills the panel from top to bottom with small headroom):

Left vertical panel — Full body, FRONT view: standing straight, natural relaxed confident pose, arms relaxed at sides, looking at camera.
Right vertical panel — Full body, BACK view: exact same standing pose seen from directly behind, same scale and same vertical alignment as the front view, showing the back of the outfit and hair.
RIGHT HALF of the canvas (occupies the other 50% of the width, full height) — split into a 2x2 GRID of 4 EQUAL PANELS (close-up portraits, head and shoulders, tight framing on the face):
3) Top-left panel — Face CLOSE-UP, FRONT view: head and shoulders, looking straight at camera, neutral calm expression.
4) Top-right panel — Head CLOSE-UP, BACK view: directly from behind, showing the back of the head and hairstyle in full detail.
5) Bottom-left panel — Face CLOSE-UP, THREE-QUARTER view from the RIGHT side (face turned slightly so we see more of the right side of the face).
6) Bottom-right panel — Face CLOSE-UP, THREE-QUARTER view from the LEFT side (face turned slightly so we see more of the left side of the face).

All 6 shots must show the SAME person, SAME outfit, SAME accessories, SAME hairstyle, SAME lighting, SAME background — only the camera angle and framing changes. Consistent character design across all frames, like an official movie character turnaround or a fashion lookbook.

Layout rules: keep the 16:9 master frame; left side = 2 vertical full-body panels of equal width and full height; right side = 2x2 grid of equal close-up panels; thin clean separators between all panels; no text, no labels, no watermarks, no logos, no extra characters, no duplicates of the person inside a single panel.

Ultra-detailed, photorealistic skin texture, realistic fabric texture, realistic hair strands, natural shadows. The final result must look like six real photographs of the same real person taken in one studio session, assembled into one professional 16:9 character reference sheet.

@Image1 sets the location, entrance, architecture, lighting and time of day — match its exact color temperature, brightness and mood. @Image2 is the rider — take the person's full appearance, face, body type, hairstyle and outfit entirely from @Image2, keep them identical throughout. POV vertical phone video, shot handheld by a person standing near the entrance filming everything on a smartphone. One continuous unbroken take, no cuts, no edits, no shot changes — single fluid shot from start to finish. Authentic smartphone footage look: slight natural hand shake, minor wobble, vertical 9:16 framing, occasional glimpse of the filmer's own hand and phone edge entering the bottom of the frame, realistic phone camera dynamic range and slight motion blur. The person from @Image2 rides up slowly on a real modern high-end sport motorcycle — black bodywork with yellow accents, realistic production motorcycle, NOT futuristic, just sleek and contemporary like a real superbike. They roll up toward the entrance of @Image1 and stop, both feet planting on the ground, confident posture. The phone follows them smoothly. They extend an index finger and press a small button on top of the fuel tank. Then, in the same continuous shot, the motorcycle smoothly folds and collapses in on itself with precise mechanical motion — wheels retract, panels rotate and compress — reassembling into a compact black-and-yellow hard-shell trolley suitcase with four wheels standing on the ground. The filmer's hand reacts in surprise. The person from @Image2 grips and pulls the telescopic handle up out of the suitcase, turns, and walks confidently toward the entrance door of @Image1, rolling the suitcase beside them. The phone keeps following them in one motion as they reach the door and step inside. Adaptive lighting: if @Image1 is dark night, soft phone flash and ambient glow; if bright daylight, natural soft light. Consistent face, body and outfit from @Image2 throughout, no distortion, no morphing. Suitcase keeps the same black-and-yellow color scheme as the motorcycle. Realistic everyday smartphone video texture, 100% real-life shooting look, no cinematic polish, one seamless continuous take throughout. Audio: real ambient street sound, motorcycle engine hum fading as they stop, soft mechanical click on button press, layered transformation servo and metallic clicks, suitcase wheels rolling, footsteps, surprised bystander reactions.

@Image1 sets the location, entrance, architecture, lighting and time of day — match its exact color temperature, brightness and mood. POV vertical phone video, shot handheld by a person standing near the entrance filming everything on a smartphone. One continuous unbroken take, no cuts, no edits, no shot changes — single fluid shot from start to finish. Authentic smartphone footage look: slight natural hand shake, minor wobble, vertical 9:16 framing, realistic phone camera dynamic range and slight motion blur. The filmer's hand, finger or phone must NEVER be visible in frame — keep the bottom of the frame completely clean, no hand, no finger, no phone edge anywhere in the shot. A stunning fit young woman with an hourglass figure, full bust, narrow waist and curvy rear, long dark wavy hair, dark sunglasses, wearing a white halter crop top and beige denim micro shorts with tan strappy sandals, rides up slowly on a real modern high-end ELECTRIC sport motorcycle — black bodywork with yellow accents, realistic production motorcycle, NOT futuristic, just sleek and contemporary like a real electric superbike, no exhaust pipes. She rolls up toward the entrance of @Image1 and stops, both feet planting on the ground, confident posture, hair settling. The phone follows her smoothly. She extends her index finger and presses a small button on top of the fuel tank. Then, in the same continuous shot, the motorcycle performs a SLOW STEP-BY-STEP MECHANICAL TRANSFORMATION like a Transformers robot — NOT a morph, NOT a quick blend. Individual metal panels detach and slide out one by one, parts rotate on visible hinges and pivots, the wheels fold inward and tuck away, the frame contracts section by section, exposed gears, pistons and hydraulic joints move with mechanical precision, plates rearrange and lock together piece by piece in a clear sequence, gradually reassembling step by step into a compact black-and-yellow hard-shell trolley suitcase with four wheels standing on the ground. Every stage of the folding is visible and physically mechanical. The woman grips and pulls the telescopic handle up out of the suitcase, turns, and walks confidently toward the entrance door of @Image1, rolling the suitcase beside her, sunglasses on. The phone keeps following her in one motion as she reaches the door and steps inside. Adaptive lighting: if @Image1 is dark night, soft phone flash and ambient glow; if bright daylight, natural soft light. Consistent facial features, body and white-top-beige-shorts outfit throughout, no distortion, no morphing of the person. Suitcase keeps the same black-and-yellow color scheme as the motorcycle. Realistic everyday smartphone video texture, 100% real-life shooting look, no cinematic polish, one seamless continuous take throughout. Audio: real ambient street sound, smooth electric motor whine and high-pitched EV hum fading as she stops (no gasoline engine, no exhaust roar), soft mechanical click on button press, detailed layered transformation sounds — servo motors whirring, metal panels sliding and locking, gears turning, hydraulic hisses, ratcheting clicks in sequence, final clunk as it locks into suitcase form, then suitcase wheels rolling, footsteps, surprised bystander reactions.

@Image1 sets the location, entrance, architecture, lighting and time of day — match its exact color temperature, brightness and mood. POV vertical phone video, shot handheld by a person standing near the entrance filming everything on a smartphone. One continuous unbroken take, no cuts, no edits, no shot changes — single fluid shot from start to finish. Authentic smartphone footage look: slight natural hand shake, minor wobble, vertical 9:16 framing, occasional glimpse of the filmer's own hand and phone edge entering the bottom of the frame, realistic phone camera dynamic range and slight motion blur. A stunning fit young woman with an hourglass figure, full bust, narrow waist and curvy rear, long dark wavy hair, dark sunglasses, wearing a white halter crop top and beige denim micro shorts with tan strappy sandals, rides up slowly on a real modern high-end ELECTRIC sport motorcycle — black bodywork with yellow accents, realistic production motorcycle, NOT futuristic, just sleek and contemporary like a real electric superbike, no exhaust pipes. She rolls up toward the entrance of @Image1 and stops, both feet planting on the ground, confident posture, hair settling. The phone follows her smoothly. She extends her index finger and presses a small button on top of the fuel tank. Then, in the same continuous shot, the motorcycle performs a SLOW STEP-BY-STEP MECHANICAL TRANSFORMATION like a Transformers robot — NOT a morph, NOT a quick blend. Individual metal panels detach and slide out one by one, parts rotate on visible hinges and pivots, the wheels fold inward and tuck away, the frame contracts section by section, exposed gears, pistons and hydraulic joints move with mechanical precision, plates rearrange and lock together piece by piece in a clear sequence, gradually reassembling step by step into a compact black-and-yellow hard-shell trolley suitcase with four wheels standing on the ground. Every stage of the folding is visible and physically mechanical. The filmer's hand reacts in surprise. The woman grips and pulls the telescopic handle up out of the suitcase, turns, and walks confidently toward the entrance door of @Image1, rolling the suitcase beside her, sunglasses on. The phone keeps following her in one motion as she reaches the door and steps inside. Adaptive lighting: if @Image1 is dark night, soft phone flash and ambient glow; if bright daylight, natural soft light. Consistent facial features, body and white-top-beige-shorts outfit throughout, no distortion, no morphing of the person. Suitcase keeps the same black-and-yellow color scheme as the motorcycle. Realistic everyday smartphone video texture, 100% real-life shooting look, no cinematic polish, one seamless continuous take throughout. Audio: real ambient street sound, smooth electric motor whine and high-pitched EV hum fading as she stops (no gasoline engine, no exhaust roar), soft mechanical click on button press, detailed layered transformation sounds — servo motors whirring, metal panels sliding and locking, gears turning, hydraulic hisses, ratcheting clicks in sequence, final clunk as it locks into suitcase form, then suitcase wheels rolling, footsteps, surprised bystander reactions.