D2C Ads Audit Checklist
Find exactly what's killing your ROAS in 5 minutes. 5-section framework covering Campaign Structure, Targeting, Creatives, Funnel, and Tracking.
Free Resources · D2C Growth · Performance Marketing
Steal the exact frameworks, checklists & systems I use to scale brands — diagnose your ad account in under 5 minutes and fix what's killing your ROAS.
Prompt Library
Reel Prompts
The exact 4-frame, 3-clip AI prompt behind the banana chips process reel — from raw banana to sealed pack, copy-paste ready.
Ingredients to hand-pleated modak, steamed and boxed, then a frozen 360° ghee-pour hero shot. The full 3-clip prompt for this Ganesh Chaturthi reel.
A confident walk-and-talk, a real sidewalk collision, and a "Comment AI" message revealed in extreme slow motion. The full 3-clip prompt behind this lead-gen reel.
A product-locked storyboard, a 22s multi-item video with holographic labels, plus standalone earring, ring and bracelet follow-ups. The full prompt system behind this luxury jewellery campaign.
A crate of energy drink sealed in ice, dropped into an active volcano, and caught mid-air out of the blast. 5 reference images plus one single continuous-shot video prompt, ready to test yourself.
A single seed falls off a pickup truck, grows into a sunflower, transforms into golden oil, and fries up into poori and bhajji. 8 reference images plus two stitchable 10s video prompts, ready to test yourself.
Every ingredient hangs frozen in mid-air — rose syrup, sabja, sev, milk, pista ice cream, pistachio — and a single touch pours each one into the glass in the exact order a real faluda is built. 8 reference images plus two stitchable 10s video prompts, ready to test yourself.
A chocolate square drops and blooms into a perfect rangoli, then a molten pour, a macro snap, a Diwali gift, and a zoom into a city billboard that loops straight back to frame one. 8 reference images plus six shot-by-shot video prompts, ready to test yourself.
One potato drops from a giant hopper, gets washed, rolls down an amber steel corridor, is wire-cut into slices, fried, caught in a masala storm and lands in a bag on a city street. 13 reference images plus ten shot-by-shot video prompts, ready to test yourself.
A phone lies upside down on a wine glass like a lid, and every fingertip tap drops one ingredient inside — ice, juice, a stir — before the phone lifts away, a watermelon slice is placed on the rim, and the camera rotates 360° around the finished drink. 7 reference images plus 8 short video prompts, ready to test yourself.
A giant maroon-and-gold mithai box rests on a lit-up Diwali street under a gold cloth. A helicopter flies in, hooks the cloth and lifts it away, then the lid opens in a burst of golden light, sweets and fireworks. 5 reference images plus short, stitchable video prompts, ready to test yourself.
A six-carriage glass hyperloop train packed with mithai races over a river town at dusk, slows down to show the sweets inside, and pulls into a marigold-lit station where a hostel student far from home receives his Diwali box. 5 reference images plus 5 short video prompts, ready to test yourself.
Jewellery Ad Prompt
The exact 2-step prompt system behind a premium luxury jewellery campaign — one product-locked storyboard, a full multi-item video, plus standalone earring, ring, and bracelet follow-ups. Copy any card and paste it into your tools.
Create a premium 16:9 HORIZONTAL PRODUCT STORYBOARD SHEET for a luxury jewellery commercial using the uploaded jewellery image as the ONLY official product reference. PRODUCT LOCK — MANDATORY: Reproduce the exact jewellery set from the reference. Maintain 100% identical: necklace design, necklace shape and proportions, earrings, pendant, bangles, rings, gold/silver metal finish, gemstones, stone colors, pearl details, stone placement, patterns, engravings, ornaments, number of jewellery pieces. DO NOT redesign, simplify, replace or invent jewellery. DO NOT change gemstone colors or placement. The uploaded image is the absolute product reference. MODEL: Include the same elegant adult Indian female jewellery model throughout. Royal Indian bridal/fashion appearance, sophisticated makeup, elegant hairstyle, premium traditional outfit, graceful expressions. Create ONE 16:9 storyboard sheet containing 8 cinematic frames: 1. HERO — model wearing the complete jewellery set 2. NECKLACE — dramatic close-up of the necklace 3. EARRINGS — macro close-up with realistic gemstone sparkle 4. PENDANT — detailed cinematic product shot 5. BANGLES & RINGS — elegant hand close-up 6. MODEL — graceful bridal fashion portrait 7. FULL JEWELLERY — complete set displayed on a luxury jewellery stand 8. FINAL HERO — model + jewellery in a grand royal environment VISUAL STYLE: Luxury Indian jewellery campaign, royal, elegant, premium wedding advertisement, cinematic fashion photography, photorealistic, realistic gemstones, realistic gold reflections, sophisticated Indian bridal aesthetic, editorial luxury photography. ENVIRONMENT: Premium Indian palace-inspired interior, subtle warm lighting, elegant marble or dark luxury background, tasteful golden highlights, soft atmospheric depth. CAMERA: 85mm fashion photography, 50mm product photography, macro lens, shallow depth of field, cinematic focus pulls, controlled reflections. IMPORTANT: Every frame must show the SAME jewellery. Maintain exact product consistency, model consistency, wardrobe consistency and lighting continuity. No unnecessary text. No extra jewellery. No duplicate pieces. No distorted hands. No malformed gemstones. No altered product design. 16:9 horizontal, 4K, ultra realistic, premium commercial storyboard.
22-second multi-item version: each piece gets its own macro beat with a holographic name label, closing on a DigitalRushi Jewellery brand card.
Create a premium cinematic jewellery commercial using the uploaded storyboard as the visual reference. PRODUCT LOCK: Keep the exact jewellery from the reference throughout the entire video — the necklace, earrings, ring, and bracelet must each appear exactly as shown in the storyboard, with no redesign, simplification, or invented details. Preserve 100%: design, shape, proportions, gemstones, colors, metal finish, pearls, patterns and every jewellery detail, for every individual piece. DO NOT redesign or alter any piece of jewellery. MODEL LOCK: Use the same adult Indian female model from the reference. Maintain identical face, hairstyle, makeup, outfit and appearance throughout. HOLOGRAPHIC LABEL STYLE (applies to every labeled shot below): A thin, elegant, glowing gold-outlined holographic label, minimal serif typeface, appears beside each jewellery piece as it's shown — not a bold tech-style overlay, a refined, jewellery-boutique-appropriate glow. Each label fades in with a subtle shimmer as the piece comes into focus, holds long enough to read clearly, then fades out cleanly before the next item appears. Only one label on screen at a time. Text must render crisp, legible, correctly spelled, and undistorted. 22-SECOND LUXURY JEWELLERY AD: 0-2s: Elegant Indian woman in a luxurious royal environment. Slow cinematic push-in. Soft golden light. No label yet. 2-5s: Extreme macro shot of the necklace. Camera slowly moves across gemstones and metalwork, realistic sparkle. Holographic label fades in beside the necklace reading: "NECKLACE" 5-8s: Macro close-up on the earrings, catching realistic gemstone sparkle from a subtle head turn. Holographic label fades in beside the earring reading: "EARRINGS" 8-11s: Elegant hand close-up focused on the ring, fingers gracefully posed. Holographic label fades in beside the ring reading: "RING" 11-14s: Elegant hand/wrist close-up focused on the bracelet, subtle wrist movement catching the light. Holographic label fades in beside the bracelet reading: "BRACELET" 14-18s: Model gracefully wearing the complete jewellery set, slowly turning toward camera. Subtle movement creates natural sparkle across all four pieces together. Elegant bridal expression. No label during this beat — this is the "full look" moment, not an individual item callout. 18-20s: Final hero shot. Model wearing the complete jewellery set in the grand luxury environment. Slow cinematic camera orbit. All four pieces remain clearly visible. 20-22s: Cut to a clean, elegant closing card: the complete jewellery set displayed on the luxury jewellery stand from the storyboard, softly lit. A refined holographic-style brand mark fades in: "DIGITALRUSHI JEWELLERY" in a minimal gold serif wordmark, with a thin decorative line beneath it. Hold on this final card as the ad ends. STYLE: Luxury Indian jewellery campaign, royal wedding advertisement, premium fashion film, cinematic lighting, realistic gold, realistic gemstones, elegant, sophisticated, 4K, 16:9, 24fps. NO: altered jewellery, extra jewellery, duplicate pieces, fake gemstones, distorted hands, extra fingers, changing face, changing outfit, product redesign, CGI-looking metal, excessive sparkle, camera shake, more than one holographic label visible at once, misspelled or distorted label text.
Same model, same environment, same holographic-label style — narrowed to one piece at a time so each also stands alone as its own 13-second ad.
Create a premium 16:9 HORIZONTAL PRODUCT STORYBOARD SHEET for a luxury jewellery commercial, focused only on the earrings, using the uploaded earring image as the ONLY official product reference. PRODUCT LOCK — MANDATORY: Reproduce the exact earring design from the reference. Maintain 100% identical: earring design and shape, metal finish, gemstones and stone colors, pearl details (if present), stone placement, patterns and engravings, drop length and proportions. DO NOT redesign, simplify, replace or invent the earring design. DO NOT change gemstone colors or placement. MODEL: Use the same elegant adult Indian female jewellery model from the necklace campaign — identical face, hairstyle, makeup, and premium traditional outfit, so this reads as part of the same campaign. She should wear ONLY the earrings in every frame — no necklace, ring, or bracelet visible. Create ONE 16:9 storyboard sheet containing 4 cinematic frames: 1. HERO — model in three-quarter profile with hair swept back, earring clearly visible and in focus 2. MACRO — extreme close-up of the earring, realistic gemstone sparkle and metalwork detail 3. ALTERNATE ANGLE — model with a subtle head turn, showing the earring's drop and movement catching the light 4. FINAL HERO — the earring alone, displayed on a small luxury velvet cushion or jewellery stand VISUAL STYLE: Luxury Indian jewellery campaign, royal, elegant, premium wedding advertisement, cinematic fashion photography, photorealistic, realistic gemstones, realistic gold reflections, sophisticated Indian bridal aesthetic, editorial luxury photography. ENVIRONMENT: Same premium Indian palace-inspired interior, subtle warm lighting, elegant marble or dark luxury background as the necklace campaign — this should feel like it belongs to the same shoot. CAMERA: 85mm fashion photography, 50mm product photography, macro lens, shallow depth of field, cinematic focus pulls, controlled reflections. No unnecessary text. No other jewellery pieces visible. No duplicate pieces. No distorted hands. No malformed gemstones. No altered product design. 16:9 horizontal, 4K, ultra realistic, premium commercial storyboard.
Create a premium cinematic jewellery commercial using the uploaded earring storyboard as the visual reference. This ad is focused only on the earrings. PRODUCT LOCK: Keep the exact earring design from the reference throughout. Preserve 100%: design, shape, gemstones, colors, metal finish, pearls, patterns and every earring detail. DO NOT redesign or alter the earring. The model should wear ONLY the earrings — no necklace, ring, or bracelet visible in any shot. MODEL LOCK: Use the same adult Indian female model from the necklace campaign. Maintain identical face, hairstyle, makeup, outfit and appearance. HOLOGRAPHIC LABEL STYLE: A thin, elegant, glowing gold-outlined holographic label, minimal serif typeface, matching the style used in the necklace ad. Fades in with a subtle shimmer, holds long enough to read, fades out cleanly. Text must render crisp, legible, correctly spelled, undistorted. 13-SECOND EARRING AD: 0-2s: Elegant Indian woman in the luxurious royal environment, hair swept back to reveal the earring. Slow cinematic push-in. Soft golden light. No label yet. 2-5s: Extreme macro shot of the earring. Camera slowly moves across gemstones and metalwork. Realistic gemstone sparkle. Holographic label fades in beside the earring reading: "EARRINGS" 5-8s: Model makes a subtle, graceful head turn, the earring swaying gently and catching the light from a new angle. Label has faded out by this point — this beat is about movement and sparkle, not the callout. 8-11s: Final hero shot. The earring displayed alone on a small luxury velvet cushion or jewellery stand, softly lit. Slow cinematic camera orbit around it. 11-13s: Cut to a clean, elegant closing card: a refined holographic-style brand mark fades in reading "DIGITALRUSHI JEWELLERY" in a minimal gold serif wordmark, with a thin decorative line beneath it. Hold on this final card as the ad ends. STYLE: Luxury Indian jewellery campaign, royal wedding advertisement, premium fashion film, cinematic lighting, realistic gold, realistic gemstones, elegant, sophisticated, 4K, 16:9, 24fps. NO: altered earring design, extra jewellery visible, duplicate pieces, fake gemstones, distorted hands, extra fingers, changing face, changing outfit, product redesign, CGI-looking metal, excessive sparkle, camera shake, misspelled or distorted label text.
Create a premium 16:9 HORIZONTAL PRODUCT STORYBOARD SHEET for a luxury jewellery commercial, focused only on the ring, using the uploaded ring image as the ONLY official product reference. PRODUCT LOCK — MANDATORY: Reproduce the exact ring design from the reference. Maintain 100% identical: ring design and band shape, metal finish, gemstones and stone colors, stone placement and setting style, patterns and engravings, proportions. DO NOT redesign, simplify, replace or invent the ring design. DO NOT change gemstone colors or placement. MODEL: Use the same elegant adult Indian female jewellery model from the earlier campaign — identical hand and skin tone, same premium traditional outfit visible at the wrist/sleeve. Only the ring should be visible on the hand — no bracelet or other rings. Create ONE 16:9 storyboard sheet containing 4 cinematic frames: 1. HERO — elegant hand pose with fingers gracefully arranged, ring clearly visible, soft luxury background 2. MACRO — extreme close-up of the ring, realistic gemstone sparkle and metalwork/setting detail 3. ALTERNATE ANGLE — hand gently rotating or fingers lightly moving, showing the ring catching light from a different facet angle 4. FINAL HERO — the ring alone, displayed on a small luxury velvet cushion or ring stand VISUAL STYLE: Luxury Indian jewellery campaign, royal, elegant, premium wedding advertisement, cinematic fashion photography, photorealistic, realistic gemstones, realistic gold reflections, sophisticated Indian bridal aesthetic, editorial luxury photography. ENVIRONMENT: Same premium Indian palace-inspired interior, subtle warm lighting, elegant marble or dark luxury background as the earlier campaigns — this should feel like it belongs to the same shoot. CAMERA: 50mm product photography, macro lens, shallow depth of field, cinematic focus pulls, controlled reflections. No unnecessary text. No other jewellery pieces visible. No duplicate pieces. No distorted hands. No extra or malformed fingers. No malformed gemstones. No altered product design. 16:9 horizontal, 4K, ultra realistic, premium commercial storyboard.
Create a premium cinematic jewellery commercial using the uploaded ring storyboard as the visual reference. This ad is focused only on the ring. PRODUCT LOCK: Keep the exact ring design from the reference throughout. Preserve 100%: design, band shape, gemstones, colors, metal finish, setting style, patterns and every ring detail. DO NOT redesign or alter the ring. Only the ring should be visible on the hand — no bracelet or other rings in any shot. MODEL LOCK: Use the same adult Indian female model's hand from the earlier campaign. Maintain identical skin tone, hand proportions, and outfit/sleeve detail visible at the wrist. HOLOGRAPHIC LABEL STYLE: A thin, elegant, glowing gold-outlined holographic label, minimal serif typeface, matching the style used in the earlier ads. Fades in with a subtle shimmer, holds long enough to read, fades out cleanly. Text must render crisp, legible, correctly spelled, undistorted. 13-SECOND RING AD: 0-2s: Elegant hand pose in the luxurious royal environment, fingers gracefully arranged, ring clearly visible. Slow cinematic push-in. Soft golden light. No label yet. 2-5s: Extreme macro shot of the ring. Camera slowly moves across the gemstone and setting detail. Realistic gemstone sparkle. Holographic label fades in beside the ring reading: "RING" 5-8s: The hand gently rotates, fingers lightly shifting, the ring catching light from a new facet angle. Label has faded out by this point — this beat is about movement and sparkle, not the callout. 8-11s: Final hero shot. The ring displayed alone on a small luxury velvet cushion or ring stand, softly lit. Slow cinematic camera orbit around it. 11-13s: Cut to a clean, elegant closing card: a refined holographic-style brand mark fades in reading "DIGITALRUSHI JEWELLERY" in a minimal gold serif wordmark, with a thin decorative line beneath it. Hold on this final card as the ad ends. STYLE: Luxury Indian jewellery campaign, royal wedding advertisement, premium fashion film, cinematic lighting, realistic gold, realistic gemstones, elegant, sophisticated, 4K, 16:9, 24fps. NO: altered ring design, extra jewellery visible, duplicate pieces, fake gemstones, distorted hands, extra or missing fingers, changing skin tone, changing outfit, product redesign, CGI-looking metal, excessive sparkle, camera shake, misspelled or distorted label text.
Create a premium 16:9 HORIZONTAL PRODUCT STORYBOARD SHEET for a luxury jewellery commercial, focused only on the bracelet, using the uploaded bracelet image as the ONLY official product reference. PRODUCT LOCK — MANDATORY: Reproduce the exact bracelet design from the reference. Maintain 100% identical: bracelet design and link/chain structure, metal finish, gemstones and stone colors, pearl details (if present), stone placement and setting style, patterns and engravings, clasp design and proportions. DO NOT redesign, simplify, replace or invent the bracelet design. DO NOT change gemstone colors or placement. MODEL: Use the same elegant adult Indian female jewellery model's hand and wrist from the earlier campaign — identical skin tone and wardrobe/sleeve detail. Only the bracelet should be visible on the wrist — no rings or other pieces. Create ONE 16:9 storyboard sheet containing 4 cinematic frames: 1. HERO — elegant wrist and forearm pose, bracelet clearly visible, soft luxury background 2. MACRO — extreme close-up of the bracelet, realistic gemstone sparkle, link/chain and clasp detail 3. ALTERNATE ANGLE — wrist gently turning or arm lightly moving, bracelet catching light from a different facet angle 4. FINAL HERO — the bracelet alone, coiled or displayed on a small luxury velvet cushion or jewellery stand VISUAL STYLE: Luxury Indian jewellery campaign, royal, elegant, premium wedding advertisement, cinematic fashion photography, photorealistic, realistic gemstones, realistic gold reflections, sophisticated Indian bridal aesthetic, editorial luxury photography. ENVIRONMENT: Same premium Indian palace-inspired interior, subtle warm lighting, elegant marble or dark luxury background as the earlier campaigns — this should feel like it belongs to the same shoot. CAMERA: 50mm product photography, macro lens, shallow depth of field, cinematic focus pulls, controlled reflections. No unnecessary text. No other jewellery pieces visible. No duplicate pieces. No distorted hands. No extra or malformed fingers. No malformed gemstones. No altered product design. 16:9 horizontal, 4K, ultra realistic, premium commercial storyboard.
Create a premium cinematic jewellery commercial using the uploaded bracelet storyboard as the visual reference. This ad is focused only on the bracelet. PRODUCT LOCK: Keep the exact bracelet design from the reference throughout. Preserve 100%: design, link/chain structure, gemstones, colors, metal finish, clasp design, patterns and every bracelet detail. DO NOT redesign or alter the bracelet. Only the bracelet should be visible on the wrist — no rings or other jewellery in any shot. MODEL LOCK: Use the same adult Indian female model's hand and wrist from the earlier campaign. Maintain identical skin tone, wrist proportions, and outfit/sleeve detail. HOLOGRAPHIC LABEL STYLE: A thin, elegant, glowing gold-outlined holographic label, minimal serif typeface, matching the style used in the earlier ads. Fades in with a subtle shimmer, holds long enough to read, fades out cleanly. Text must render crisp, legible, correctly spelled, undistorted. 13-SECOND BRACELET AD: 0-2s: Elegant wrist and forearm pose in the luxurious royal environment, bracelet clearly visible. Slow cinematic push-in. Soft golden light. No label yet. 2-5s: Extreme macro shot of the bracelet. Camera slowly moves across the links, clasp, and gemstone detail. Realistic gemstone sparkle. Holographic label fades in beside the bracelet reading: "BRACELET" 5-8s: The wrist gently turns, arm lightly shifting, the bracelet catching light from a new facet angle, links shifting naturally with the movement. Label has faded out by this point — this beat is about movement and sparkle, not the callout. 8-11s: Final hero shot. The bracelet displayed alone, elegantly coiled on a small luxury velvet cushion or jewellery stand, softly lit. Slow cinematic camera orbit around it. 11-13s: Cut to a clean, elegant closing card: a refined holographic-style brand mark fades in reading "DIGITALRUSHI JEWELLERY" in a minimal gold serif wordmark, with a thin decorative line beneath it. Hold on this final card as the ad ends. STYLE: Luxury Indian jewellery campaign, royal wedding advertisement, premium fashion film, cinematic lighting, realistic gold, realistic gemstones, elegant, sophisticated, 4K, 16:9, 24fps. NO: altered bracelet design, extra jewellery visible, duplicate pieces, fake gemstones, distorted hands, extra or missing fingers, changing skin tone, changing outfit, product redesign, CGI-looking metal, excessive sparkle, camera shake, misspelled or distorted label text.
Try It Yourself
The exact 5-image + single continuous-shot video prompt behind our "dropped a case of energy drink into an active volcano" reel. Copy any card and paste it into your own tools.
A photorealistic wide shot inside the open cargo bay of a military-style transport plane, ramp fully lowered, an erupting volcano and glowing crater visible in the distance through the opening. In the foreground, a wooden crate packed with DigitalRushi Energy cans (matte graphite-black cans with a bold electric-blue lightning-bolt logo and "DIGITALRUSHI ENERGY" wordmark, clearly legible) sits fully encased inside a massive, perfectly clear block of ice on the cargo bay's roller track, condensation-free, ice showing fine internal cracks and refracted light. Warm interior cargo-bay light mixes with a distant orange volcanic glow. Composition: the ice-encased crate fills the foreground and mid-ground, the volcano framed through the ramp opening behind it. Lens: 24mm wide-angle equivalent, deep focus keeping both the crate and the distant volcano sharp. Color palette: cool blue-white ice and graphite-black cans against a warm orange-gray volcanic horizon. Photorealistic, cinematic commercial aesthetic. Aspect ratio 9:16.
A photorealistic image in the same cargo bay setting as Image A, the ice-encased crate now at the very edge of the open ramp, tipping forward, roller track visible beneath it, the volcano's crater now directly visible below through the opening, glowing lava faintly at its base. Lighting: same warm interior light meeting the volcanic glow, slightly brighter now as the ramp opening dominates more of the frame. Camera: same 24mm wide angle, slightly lower to emphasize the block teetering at the edge. Color palette: consistent with Image A. Photorealistic, ultra-detailed ice and metal-crate texture. Aspect ratio 9:16.
A photorealistic shot looking down and outward from the plane's ramp, the ice-encased crate now mid-fall, tumbling slightly, smoke and ash swirling around it, the volcano's crater directly beneath it filling the lower half of frame, glowing lava visible at the crater's base through drifting steam. The ice block is sharp and in focus, motion-streaked smoke trailing above it. Lighting: dramatic underlighting from the crater's glow on the block's underside. Camera: looking down at a steep angle, 28mm equivalent, slight motion blur on background smoke only. Color palette: deep volcanic orange and black against the cool white ice block. Photorealistic, high-drama aesthetic. Aspect ratio 9:16.
A photorealistic, dramatic low-angle shot from the crater's rim looking down and across at the exact instant the ice block contacts the crater floor — a colossal eruption of lava, fire, and steam bursts upward around the impact point, thick black smoke billowing. At the center of the blast, the ice has already shattered into scattering chunks, and a single DigitalRushi Energy can is visible mid-launch, rocketing upward and outward on the shockwave, logo still clearly legible and undamaged, trailing a thin mist of ice-cold condensation. Lighting: blinding orange-white blast core fading to deep red-black smoke at the edges. Camera: low angle looking up and across, 24mm wide equivalent, capturing the blast's full scale with the can as the sharp focal point against the chaos. Color palette: white-hot orange core, black smoke, the graphite-black can with its blue logo as the one clean, sharp element in frame. Photorealistic, blockbuster-commercial aesthetic. Aspect ratio 9:16.
A photorealistic shot of an athletic young man in technical outdoor/adventure gear (fitted jacket, small chest-mounted camera, sunglasses pushed up) standing on a rocky volcanic ledge at the crater's rim, one arm extended upward, hand closing around the DigitalRushi Energy can at the exact moment of the catch, small ice shards and mist scattering away from his hand in the impact, an intense but controlled expression. The volcano's smoke and glow fill the background, slightly out of focus. The can is razor-sharp and fully legible in his grip — matte graphite-black body, electric-blue lightning-bolt logo, condensation beading on its surface from the ice. Lighting: warm volcanic backlight rim-lighting his silhouette, cooler fill light on his face and the can itself so both stay clearly visible. Camera: low-angle hero shot, 50mm equivalent, shallow depth of field with the can and his hand in sharp focus, volcano softly blurred behind. Color palette: warm orange background, cool blue-white product accent — the can's blue logo should visually pop against the warm backdrop. Photorealistic, premium energy-drink commercial aesthetic. Aspect ratio 9:16.
One unbroken ~18-second take that moves through all 5 images without cutting — built for Google Flow's Omni Flash multi-reference mode. Upload Images A-E as references, in that order, alongside this prompt.
A single unbroken continuous shot, no cuts, moving through five connected moments in one flowing take. Starting composition: exactly Reference Image A — inside the open cargo bay of a transport plane, a wooden crate of DigitalRushi Energy cans sealed inside a massive clear ice block sits on the ramp's roller track, an erupting volcano visible through the open ramp behind it. Full continuous motion, in order: crew members push the ice-encased crate smoothly along the roller track toward the open ramp (0-3s), the block tipping and teetering right at the edge exactly as in Reference Image B (3-4s); it tips forward and begins falling immediately, tumbling gently end-over-end as the camera tracks downward with it, the plane's ramp shrinking away above as the crater grows larger and more detailed below, smoke thickening around the block, passing through the composition of Reference Image C mid-fall (4-8s); the block continues its fall and makes contact with the crater floor, the instant of contact triggering a massive, sudden volcanic eruption — lava and fire bursting upward and outward in a colossal fountain, the shockwave visibly shattering the ice into scattering fragments, with a single DigitalRushi Energy can thrown clear of the blast, rocketing upward and outward on the force of the explosion, matching Reference Image D's framing as the can clears the smoke, sharp and undamaged (8-12s); the camera follows the can's arc away from the crater as the smoke and fire fall behind it, revealing an athlete standing on a rocky ledge at the crater's rim already reaching upward in anticipation, tracking the can's final approach until his hand closes around it in one clean, confident motion, ice shards and fine mist scattering outward from the catch, settling into Reference Image E's exact hero framing as he pulls the can into a steady grip, fully intact and undamaged (12-16s); camera settles into a smooth, locked-off low-angle hero hold on the can and the volcano glowing softly out of focus behind him (16-18s). Camera: begins static/wide inside the cargo bay, transitions into a smooth downward tracking shot matching the block's fall speed, holds through the eruption impact, whip-tilts upward to follow the can's launch, then follows its arc across to the catch before settling into a locked-off low-angle push-in for the final hero hold — one continuous camera performance, no hard cuts anywhere in the sequence. Lighting continuity: warm cargo-bay interior tone at the start, shifting gradually to harsher saturated orange volcanic light during the fall, a sudden blinding orange-white flash exactly at the moment of impact, settling into a sustained bright blast-glow through the launch, then transitioning to a warm rim-light/cool-fill balance on the athlete for the final catch and hero hold. Light should evolve continuously throughout — no jarring, unmotivated lighting jumps between beats. Physics: realistic weight and momentum throughout — a heavy block wobbling before it tips, real tumbling mass during the fall, a genuinely powerful and violent-looking eruption with debris, fire, and force at impact, a can trajectory that clearly reads as being thrown by that exact explosive force, and a fast, confident, believable one-handed catch with real ice-shard scatter — never slow-motion or floaty in a way that undercuts the athleticism. Audio (continuous, no hard resets): roller-track and distant volcanic rumble at the start, rising wind-rush during the fall, a huge deep explosive boom exactly on contact, a sustained roar through the launch, fading down as the can approaches the catch, a sharp satisfying catch sound (hand-on-can, ice shards) right on contact, then a confident musical sting as the hero shot settles. No dialogue anywhere in the sequence. On-screen text: none until the very final hold, where "Built to Survive Anything." fades in beneath the DigitalRushi Energy logo, crisp, correctly spelled, and fully legible. Ending composition: exactly Reference Image E, held for the final beat with the tagline fully visible. Aspect ratio 9:16. Motion intensity: escalates from calm to violent at the eruption, then resolves to confident and controlled by the catch — a clear energy arc across the full 18 seconds, not a flat continuous intensity.
Tip: if your tool loses coherence trying to hold all 5 beats in one generation, fall back to 4 separate clip-by-clip prompts (A→B, B→C, C→D, D→E) using the same continuity, lighting and physics details described above — same story, just generated and edited together instead of interpolated in one pass. I can't guarantee every AI video tool renders the eruption in one pass without flagging it for review — policies vary by tool.
Try It Yourself
The exact 8-image + two stitchable 10-second video prompts behind our "seed becomes sunflower becomes oil becomes food" reel. Copy any card and paste it into your own tools.
A photorealistic low-angle shot at roadside-dirt level, a single sunflower seed caught mid-air just after bouncing off the back of a moving pickup truck loaded with sacks, the truck slightly motion-blurred in the background driving away down a rural highway, dry grass and dusty soil at the shot's edges, overcast daylight. Camera: 35mm equivalent, shallow depth of field with the seed in sharp focus and the truck softly blurred. Color palette: dusty browns and greens, muted overcast sky. Photorealistic, naturalistic commercial aesthetic, no text. Aspect ratio 9:16.
A photorealistic extreme close-up, ground-level macro shot of the same patch of roadside soil, a small green sunflower sprout just breaking through the dirt with two tiny leaves unfurling, a faint warm sunlight glow on the shoot. Shallow depth of field, soil grains sharply detailed in the foreground. Color palette: rich brown soil, fresh green sprout. Photorealistic, macro nature-photography aesthetic. Aspect ratio 9:16.
A photorealistic medium shot of a single tall, fully bloomed sunflower in the same roadside location, its bright yellow petals fully open and facing camera, thick green stem, a soft blurred rural background with the highway faintly visible. Warm late-afternoon sunlight, golden-hour glow. Camera: 50mm equivalent, shallow depth of field with the flower head in crisp focus. Color palette: vivid yellow petals, deep brown seed center, green stem, warm gold ambient light. Photorealistic, vibrant commercial nature photography aesthetic. Aspect ratio 9:16.
A photorealistic close-up of the same sunflower's seed-head, now cracked open at the center with a glowing warm light emanating from within, a stream of thick golden oil visibly pouring out from the flower's core and arcing downward into a clear glass bottle positioned just beneath it, the bottle already partially filled with rich golden-yellow oil. Studio-clean softly blurred background transitioning from the field setting. Lighting: warm golden backlight through the oil stream, glossy highlights on the bottle glass. Camera: 50mm macro-leaning lens, shallow depth of field. Color palette: deep gold oil, warm amber highlights, soft green field bokeh in the background. Photorealistic, premium commercial product-photography aesthetic. Aspect ratio 9:16.
A photorealistic studio product shot of a clear glass bottle filled with golden sunflower oil, labeled "DIGITALRUSHI SUNFLOWER GOLD" in clean bold typography with a small sunflower icon on the label, positioned on a light wooden kitchen counter, a single sunflower head resting softly out of focus beside it. Lighting: soft diffused studio key light with a warm rim light catching the oil's golden color. Camera: 50mm lens, shallow depth of field, label sharp and legible. Color palette: golden oil, warm wood tones, clean neutral background. Photorealistic, premium FMCG commercial product photography. Aspect ratio 9:16.
A photorealistic close-up top-down shot of golden sunflower oil pouring from the bottle in Image E into a hot black kadhai (Indian wok) on a stovetop, the oil catching the light as it streams down, the pan's surface just beginning to shimmer with heat. Lighting: warm kitchen lighting with a highlight on the pouring oil stream. Camera: 50mm macro-leaning lens, shallow depth of field. Color palette: golden oil, dark pan, warm kitchen ambiance. Photorealistic, appetizing commercial food-photography aesthetic. Aspect ratio 9:16.
A photorealistic top-down close-up of the same hot kadhai now full of shimmering golden sunflower oil, a puffed golden-brown poori mid-fry rising and puffing up in the oil, with a few bhajji (vegetable fritters) already frying golden-brown beside it, small realistic oil bubbles and a light steam wisp visible. Lighting: warm, appetizing, with highlights on the oil's surface and the fried items' crisp texture. Camera: 50mm macro lens, shallow depth of field. Color palette: golden oil, deep golden-brown fried items. Photorealistic, indulgent commercial food photography aesthetic. Aspect ratio 9:16.
A photorealistic top-down flat-lay shot of a finished plate of crisp golden poori (puffed, hollow, lightly blistered) and bhajji (vegetable fritters, golden-brown, crisp-edged) arranged appetizingly beside the "DIGITALRUSHI SUNFLOWER GOLD" oil bottle from Image E — the bottle standing fully upright and stable on the counter, label facing camera and clearly legible, no tilt or pour motion. Both plate and bottle rest on a light wooden kitchen counter with a few loose sunflower petals scattered decoratively nearby. Baked-in graphic text: a clean bold headline centered in the lower third reading "FROM SEED TO SIZZLE." beneath the brand's logo/wordmark, small tagline beneath in lighter type: "DigitalRushi Sunflower Gold — Pure, From the Field." Text sharp, correctly spelled, legible at thumbnail size. Lighting and color consistent with Image G. Aspect ratio 9:16, photorealistic, warm premium commercial aesthetic.
Generate each clip separately, then cut them together at the 10-second mark — both use the same static camera logic and lighting continuity so the join should read as seamless.
Starting composition: exactly Image A — the seed mid-air just after bouncing off the truck. Full motion, beat by beat: 0:00-0:02, the seed falls and lands on the dirt, bouncing once and settling into the soil, camera static with slight handheld-style micro-shake, truck driving away and out of frame. 0:02-0:04, match-cut/dissolve to ground level (Image B): the seed's landing spot begins to stir, a small green sprout pushes up through the dirt in a fast, fluid time-lapse motion, two leaves unfurling. 0:04-0:06, the sprout accelerates its growth in continuous fast-motion, the stem shooting upward and thickening, a bud forming and rapidly blooming open into the full sunflower of Image C, petals unfurling outward in one fluid motion, camera pulling back smoothly to reveal the full flower. 0:06-0:08, the sunflower's center darkens and begins to glow warmly from within; the seed-head cracks open at its core (matching Image D), a thick stream of golden oil beginning to pour out and arc downward. 0:08-0:10, the oil stream continues pouring in a steady, glossy arc into a clear glass bottle rising into frame beneath it, the bottle filling toward the level shown in Image D, camera holding a slow push-in on the pour. Camera: starts static/handheld at ground level, smoothly transitions to a slow pull-back during the growth beat, then a slow push-in during the oil-pour beat — no jarring angle changes, one continuous camera logic throughout. Physics: natural bounce-and-settle for the seed, fluid organic growth curves for the plant (not robotic/linear), realistic viscous liquid behavior for the oil stream with a convincing arc and surface shimmer. Lighting continuity: overcast daylight at the start warming gradually into golden-hour tones by the bloom, shifting into a warm internal glow for the oil-crack moment — one continuous lighting mood, not an abrupt jump. Audio: a soft thud as the seed lands, ambient rustling wind and rural roadside sound under the growth beat, a rising magical shimmer tone as the flower blooms, a warm resonant tone plus a liquid-pouring sound as the oil begins to stream. No dialogue. Motion intensity: low at the hook, building to a fast, satisfying rush during the growth time-lapse, settling to a slow, glossy, deliberate pace for the oil pour. Ending composition: exactly Image D, oil mid-pour into the bottle. Aspect ratio 9:16.
Starting composition: the oil bottle now filled, transitioning to Image E's clean product hero framing on a kitchen counter. Full motion, beat by beat: 0:10-0:11, the filled bottle settles into Image E's clean studio/kitchen composition, held briefly as a clear product beat. 0:11-0:12.5, the bottle tilts and pours; oil streams out and down into a hot kadhai below (matching Image F), catching the light as it falls, the pan surface just beginning to shimmer with heat. 0:12.5-0:15, the oil surface starts to shimmer and ripple with heat; a ball of dough drops in and instantly puffs up into a golden poori, rising and ballooning in real time exactly as in Image G, with a few bhajji already frying golden-brown beside it, small bubbles and a light steam wisp visible. 0:15-0:17.5, the fried poori and bhajji lift out of the oil in one smooth motion and land plated on a waiting dish, oil glistening on their crisp golden-brown surfaces, camera pulling back slightly to reveal the fuller plate. 0:17.5-0:20, the plate of poori and bhajji settles next to the oil bottle — the bottle standing fully upright and still, label facing camera — in Image H's final flat-lay composition, sunflower petals settling decoratively around them, the brand card text ("FROM SEED TO SIZZLE." / "DigitalRushi Sunflower Gold — Pure, From the Field.") fading in cleanly on top, held through the end of the clip. Camera: static-to-slow-push-in top-down framing for the frying beats, matching Video Prompt 1's camera logic (deliberate, unhurried movement, no whip pans); a gentle pull-back at 0:15 to reveal the plated result; settling to a static locked frame for the brand card. Physics: realistic oil-pour viscosity, authentic dough-puffing behavior for the poori (rapid but physically believable ballooning), natural sizzle-bubble movement in the oil. Lighting continuity: warm kitchen lighting held constant from the end of Video Prompt 1 through to the brand card — no color-temperature jump at the 0:10 cut. Audio: a soft product-beat chime as the bottle settles, a glossy pour sound, a satisfying sizzle building through the fry, a gentle plating clink and warm musical resolve under the brand card. No dialogue. Motion intensity: calm and deliberate throughout — this half is about appetite appeal and product clarity, not fast action. Ending composition: exactly Image H, held on the brand card. Aspect ratio 9:16.
Tip: the sunflower-to-oil transformation (Image D) has no real-world physical reference, so it's the hardest single beat to render convincingly — budget the most regeneration attempts there. If it doesn't land, a solid fallback is a hard cut/dissolve straight from the full-bloom flower to the filled bottle, implying the transformation instead of showing it in detail.
Try It Yourself
The exact 8-image + two stitchable 10-second video prompts behind our "every ingredient hangs in the air, one touch pours it in" faluda reel — built in the real recipe order a faluda is actually assembled. Copy any card and paste it into your own tools.
The one rule that makes this mechanic actually work: once the hand touches an ingredient, that ingredient's suspended form must be fully gone from the air — not duplicated, not still hanging there — as it transfers into the glass. Every prompt below says this explicitly for exactly that reason.
A photorealistic studio shot against a pure matte black background, a tall, narrow, clear glass (traditional Indian faluda-glass shape) positioned center-frame on a dark reflective surface, completely empty. Arranged in the air around it, each frozen mid-motion with nothing touching the glass yet: a thick ribbon of vivid pink-red rose syrup curling to the upper-left, a small loose cluster of translucent gray-black soaked sabja (basil) seeds glistening with moisture to the upper-right, a coiled portion of white falooda vermicelli strands suspended just above the glass, a smooth stream of white milk suspended higher up, a single scoop of pale pistachio-green ice cream floating to one side, and a light scatter of finely chopped bright-green pistachio slivers floating near it. A stylized elegant hand (clean, soft-lit, fingers gently curved, no visible skin texture flaws) enters the frame from the right, fingertip reaching toward the rose syrup ribbon, not yet touching it. Lighting: single dramatic soft key light from above-left, deep black shadows, each suspended ingredient catching its own soft highlight. Camera: straight-on studio shot, 50mm equivalent, shallow depth of field, everything in crisp focus. Color palette: pure black background, vivid pink-red syrup, translucent gray sabja, white vermicelli and milk, pale pistachio green. Photorealistic, high-end dessert commercial aesthetic. Aspect ratio 9:16.
The exact same black-void studio setup as Image A, the hand's fingertip now touching the empty space where the rose syrup ribbon used to hang — the syrup ribbon is now completely gone from the air, fully transferred into the glass, where it has pooled into a vivid pink-red layer at the bottom. The space where the syrup used to float is now empty black void, nothing left hanging there. The other five ingredients (sabja, vermicelli, milk, ice cream, pistachio) remain exactly where they were, completely frozen and untouched. Lighting and camera identical to Image A. Photorealistic, premium dessert commercial aesthetic. Aspect ratio 9:16.
The exact same black-void studio setup, the rose syrup now settled as a clean pink-red layer at the bottom of the glass with empty black void where it used to hang, the hand having moved to touch the empty space where the sabja seed cluster used to float — the sabja cluster is now completely gone from the air, fully transferred into the glass, settled into an even layer on top of the syrup. The space where the sabja used to float is now empty black void. The vermicelli, milk, ice cream, and pistachio remain exactly where they were, completely frozen and untouched. Lighting and camera identical to previous images. Photorealistic, premium dessert commercial aesthetic. Aspect ratio 9:16.
The exact same black-void studio setup, the sabja seeds now settled in an even layer atop the syrup with empty black void where both used to hang, the hand having moved to touch the empty space where the falooda vermicelli used to float — the vermicelli is now completely gone from the air, fully transferred into the glass, settled naturally into a loose coil on top of the sabja layer. The space where the vermicelli used to float is now empty black void. The milk, ice cream, and pistachio remain exactly where they were, completely frozen and untouched. Lighting and camera identical to previous images. Photorealistic, premium dessert commercial aesthetic. Aspect ratio 9:16.
The exact same black-void studio setup, the vermicelli now settled on top of the sabja and syrup layers with empty black void above where all three used to hang, the hand having moved to touch the empty space where the milk stream used to float — the milk is now completely gone from the air, fully transferred into the glass, filling it and creating a soft, natural marbled gradient where the white milk meets the pink syrup layer below, the vermicelli and sabja now gently suspended within the milk. The space where the milk used to float is now empty black void. The ice cream and pistachio remain exactly where they were, completely frozen and untouched. Lighting and camera identical to previous images. Photorealistic, premium dessert commercial aesthetic. Aspect ratio 9:16.
The exact same black-void studio setup, the glass now filled with milk and its layered contents visible beneath the surface, empty black void above where the syrup, sabja, vermicelli and milk used to hang, the hand having moved to touch the empty space where the pista ice cream scoop used to float — the ice cream scoop is now completely gone from the air, fully transferred onto the glass, resting on top of the milk's surface, holding its rounded scoop shape. The space where the ice cream used to float is now empty black void. The pistachio garnish remains exactly where it was, completely frozen and untouched. Lighting and camera identical to previous images. Photorealistic, premium dessert commercial aesthetic. Aspect ratio 9:16.
The exact same black-void studio setup, the ice cream scoop now resting on top of the glass, empty black void everywhere ingredients used to hang, the hand touching the empty space where the chopped pistachio garnish used to float — the pistachio is now completely gone from the air, fully transferred onto the ice cream scoop, scattered naturally across its surface. The space where the pistachio used to float is now empty black void — nothing remains suspended in the frame at all. The hand is beginning to withdraw from frame to the right, and a long traditional faluda spoon is caught mid-drop just beside the glass. Lighting and camera identical to previous images. Photorealistic, premium dessert commercial aesthetic. Aspect ratio 9:16.
The exact same black-void studio setup, no hand visible, the finished pista faluda glass now fully dewed with realistic condensation droplets running down its sides, the long spoon resting naturally against the glass, every layer clearly visible through the glass from bottom to top: rose syrup, sabja seeds, vermicelli, milk, and the pistachio-scattered ice cream scoop on top. Baked-in graphic text: a small, elegant logo mark reading "DIGITALRUSHI" in clean modern lettering fading in beside the glass, with a thin minimal line beneath it. Lighting and camera identical to previous images. Text sharp, correctly spelled, legible at thumbnail size. Photorealistic, premium dessert commercial aesthetic. Aspect ratio 9:16.
Generate each clip separately, then cut them together at the 10-second mark — both use the same static camera logic and lighting continuity so the join should read as seamless.
Starting composition: exactly Image A — an empty tall glass in a black void, all six ingredients (syrup, sabja, vermicelli, milk, ice cream, pistachio) frozen suspended around it, the hand just entering frame and reaching toward the syrup. Full motion, beat by beat: 0:00-0:02, static hold on the full suspended tableau, every ingredient gently motionless as if weightless, the hand's fingertip approaching the syrup ribbon but not yet touching it. 0:02-0:04, the fingertip makes contact with the rose syrup ribbon; the instant it touches, the entire ribbon fully empties out of the air and travels down into the glass in one continuous motion, leaving nothing behind at its original position — that patch of air is now empty black void — as it pools into a clean pink-red layer at the bottom, matching Image B. The other five ingredients remain completely frozen and untouched during this beat. 0:04-0:06, the hand moves smoothly to the sabja seed cluster and touches it; the cluster immediately and fully vacates its suspended position — no seeds left hanging in the air — dropping and scattering gently onto the syrup layer below, settling into an even layer, matching Image C. The vermicelli, milk, ice cream, and pistachio remain frozen and untouched. 0:06-0:08, the hand moves to the suspended vermicelli and touches it; the entire coil immediately and fully vacates the air — nothing left hanging where it was — lowering smoothly into the glass and settling into a natural loose coil on top of the sabja, matching Image D. The milk, ice cream, and pistachio remain frozen and untouched. 0:08-0:10, the hand moves up to the suspended milk stream and touches it; the entire stream immediately and fully vacates the air — no milk left hanging — pouring smoothly down into the glass, filling it and creating a soft marbled gradient where the milk meets the syrup below, the vermicelli and sabja now gently suspended within it, matching Image E. The ice cream and pistachio remain frozen and untouched, waiting for their turn. Camera: a single locked-off straight-on studio framing for the entire clip, no camera movement at all — every bit of motion comes from the ingredients reacting to touch, not the camera. Physics: each ingredient must stay completely frozen and motionless until the exact moment the hand makes contact with it — that contact is the trigger, and the reaction should be immediate and direct. Critically, once triggered, an ingredient's suspended form must be fully consumed and vacated from the air as it transfers into the glass — it should never remain visibly floating in its original position after being poured. The syrup pours and pools like a real viscous liquid, the sabja seeds scatter and settle with light, natural weight, the vermicelli lowers and coils naturally like real soft noodles, the milk pours and marbles with realistic fluid dynamics against the syrup layer — and in every case, the source location in the air is left completely empty once the transfer is done. Lighting continuity: the single dramatic key light and black void background stay completely constant throughout — no lighting shifts in this clip. Audio: a soft, magical "activation" chime or shimmer sound at the exact instant of each touch, followed by the natural sound of that ingredient's motion (a syrup pour, a light scatter-and-settle sound, a soft noodle-settling sound, a smooth milk pour), a light ambient studio hush under everything, with a minimal, elegant, slightly playful musical bed building gently with each successive trigger. No dialogue, no voiceover. Motion intensity: each beat starts from complete stillness and bursts into motion only at the moment of contact — this stop-start rhythm (still, touch, react, still, touch, react) is the core visual signature of the whole concept and should feel crisp and deliberate, not blended together. Ending composition: exactly Image E, held briefly with the glass filled with milk and its layers visible, the ice cream and pistachio still waiting, untouched, in their suspended positions. Aspect ratio 9:16.
Starting composition: exactly Image E — the glass filled with milk and its layered contents visible, the ice cream scoop and pistachio garnish still frozen in their suspended positions, the hand having just finished triggering the milk. Full motion, beat by beat: 0:10-0:12, the hand moves to the suspended pista ice cream scoop and touches it; the scoop immediately and fully vacates its suspended position — nothing left floating where it was, that patch of air now empty black void — lowering smoothly to rest on top of the milk's surface, holding its rounded shape, matching Image F. 0:12-0:14, the hand moves to the suspended pistachio garnish and touches it; it immediately and fully vacates the air — no pistachio left hanging anywhere in frame — sprinkling down naturally onto the ice cream scoop, scattering across its surface, matching Image G. 0:14-0:15, the hand withdraws smoothly out of frame to the right; a long traditional faluda spoon drops from just above frame, landing naturally against the side of the glass, matching Image G's final state. 0:15-0:18, with the hand now gone from frame and the air completely empty (nothing suspended anywhere), condensation begins forming and gently beading down the glass's sides in realistic rivulets, the whole composition settling into complete stillness. 0:18-0:20, the "DIGITALRUSHI" logo mark fades in cleanly beside the glass exactly as shown in Image H, crisp and correctly spelled, held through the end of the clip. Camera: the same single locked-off straight-on studio framing as Video Prompt 1, completely unchanged — no movement at all in either clip. Physics: the ice cream scoop should lower and settle with believable soft-solid weight, holding its shape rather than melting or deforming unnaturally, and must be fully gone from its suspended position the instant it starts moving — never visible in two places at once; the pistachio garnish should scatter and fall with light, realistic weight, equally fully vacating the air; the spoon should drop and land with natural weight and a small settle-wobble; condensation droplets should bead and run with realistic gravity and surface tension. Lighting continuity: identical key light and black void background carried over exactly from the end of Video Prompt 1 — no shift in color temperature or brightness at the 0:10 cut. Audio: the same soft "activation" chime at each of the two touches in this clip, a light settling sound as the ice cream lands, a soft scatter sound for the pistachio, a gentle clink as the spoon lands, quiet ambient studio hush, the same minimal musical bed from Video Prompt 1 continuing and resolving into a clean, satisfying final note as the logo fades in. No dialogue, no voiceover. Motion intensity: two final crisp still-touch-react bursts for the ice cream and pistachio, then calm and deliberate for the rest of the clip, slowing to completely still by the final logo reveal. Ending composition: exactly Image H, held on the brand reveal. Aspect ratio 9:16.
Tip: the single most important rule in this whole concept is that a triggered ingredient must fully disappear from its suspended position, not just also appear in the glass — if a tool renders a ghost or duplicate lingering in the ingredient's original spot, try adding "the object has completely left its original position, there is no duplicate" to that specific prompt, or generate that one transition as a cut/dissolve instead of a continuous morph. This is also the most ingredient-dense version of this mechanic (six components) — if your tool can't hold all six frozen at once, a solid fallback is to drop sabja and the pistachio garnish and keep just syrup, vermicelli, milk, and ice cream.
Try It Yourself
The exact 8-image + six shot-by-shot video prompts behind our Diwali chocolate reel — a chocolate square that blooms into a rangoli, a molten pour, a macro snap, a festive gift, and a city billboard zoom that loops back to the first frame. Copy any card and paste it into your own tools.
Palette lock for every prompt: deep purple #3D1A6E, mid purple #5B2C8F, gold #C9A24B, cream #F5EFE6, milk-chocolate brown. Keep the wordmark in a bold rounded sans-serif so the bar reads as its own brand, not a copy of an existing one.
Commercial packshot of a premium milk chocolate bar in a flat rectangular flow-wrap wrapper, deep royal purple (#3D1A6E) matte wrapper with subtle satin sheen, the brand name "digitalrushi" in lowercase bold rounded geometric sans-serif, metallic gold foil (#C9A24B) lettering centered horizontally, thin gold horizontal pinstripe above and below the name, small cream-white (#F5EFE6) text "MILK CHOCOLATE" beneath, crimped sealed ends at left and right, bar lying flat at a slight 15° angle, seamless deep purple backdrop matching the wrapper, soft large top-down softbox key light with gentle gradient falloff, crisp edge highlight along wrapper crimp, sharp focus across entire product, 100mm macro lens, photorealistic commercial product photography, clean legible typography, no extra text, single product, 9:16
Perfect top-down overhead view of a symmetrical circular rangoli pattern made entirely of milk chocolate shavings, chocolate curls, cocoa powder lines and scattered edible gold foil flakes, a single glossy milk chocolate square sitting at the exact center, laid on a seamless glossy deep royal purple (#3D1A6E) floor, eight-fold radial petal design with fine detailing, soft even overhead light with subtle specular glints on the gold flakes, crisp texture of each shaving visible, shallow falloff vignette toward frame edges, photorealistic food styling, premium dessert advertising photography, 9:16 vertical
Stack of five unwrapped milk chocolate bars with neat segmented squares, stacked in a slightly offset staircase, a thick glossy ribbon of molten milk chocolate pouring from top of frame folding over the top bar and cascading down the sides, one wrapped "digitalrushi" bar (reference product, Image A) leaning upright beside the stack with logo facing camera, seamless deep purple (#3D1A6E) background, low 3/4 angle, 85mm lens, f/4 with soft background falloff, strong backlight rim making the chocolate ribbon glow, warm fill light from front-left, glossy liquid highlights, photorealistic premium chocolate commercial photography, 9:16
Young Indian woman in her early 20s with a shoulder-length black bob and blunt bangs, wearing a deep purple knit sweater with a cream and gold festive pattern, sitting at a purple table with her chin resting on one hand, joyfully squinting and smiling, holding the "digitalrushi" chocolate bar (reference product, Image A) upright covering one eye with logo facing camera, several wrapped bars scattered on the table, fully monochrome deep purple (#3D1A6E) set with walls and table in the same purple, warm string-light bokeh in gold far behind, medium close-up, eye-level, 85mm lens, f/2.8, soft beauty key light from front-right, natural skin texture, correct hands with five fingers, photorealistic lifestyle advertising photography, 9:16
Extreme macro close-up of two fingertips snapping a milk chocolate square off a bar, the break line showing a clean glossy cross-section, tiny chocolate crumbs frozen in mid-air, peeled-back purple wrapper with gold edge visible at bottom of frame, deep purple background, 100mm macro lens, f/5.6, razor-sharp focus on the break, hard side light raking across the surface texture, photorealistic macro food photography, 9:16
Two hands, one with a gold bangle and mehendi, handing a rectangular deep purple gift box tied with a gold satin ribbon across a purple table, the "digitalrushi" wordmark in gold foil on the lid, three lit clay diyas with small warm flames out of focus in the foreground, deep purple room behind with soft gold bokeh, side angle at table height, 50mm lens, f/2, warm candlelight-toned key, photorealistic festive lifestyle advertising photography, natural hand anatomy, 9:16
Wide night shot of a busy Indian city junction during Diwali, a tall vertical portrait-format (9:16) LED digital billboard mounted on the facade of a mid-rise building at the center of the frame, the billboard displaying a deep purple (#3D1A6E) chocolate advertisement with the "digitalrushi" chocolate bar (reference product, Image A) standing on a purple pedestal and gold text "digitalrushi" in lowercase bold rounded sans-serif, the screen glowing brightly and casting purple light on the surrounding wall, strings of warm golden fairy lights and lanterns across the street, light traffic with motion-blurred headlights and tail lights, a few pedestrians at street level, slightly low eye-level camera looking straight at the billboard, perfectly frontal and centered, no perspective tilt, 35mm lens, deep focus, wet road reflecting purple and gold, photorealistic cinematic night city photography, 9:16
The exact same deep purple LED billboard screen now filling the entire 9:16 frame edge to edge, perfectly frontal, displaying a top-down view of a symmetrical circular rangoli made of milk chocolate shavings and gold foil flakes with a single chocolate square at the center on a glossy deep purple floor (matching Image B), faint visible LED pixel grid texture across the image, slight screen glow, no building or street visible, photorealistic, 9:16
Generate each clip from its matching reference image, then cut them together in order. Shots 1–5 run about 2 seconds each; the billboard reveal runs the final 5 seconds and loops back to Shot 1.
Locked-off top-down camera. Start on an empty glossy deep purple floor. A single milk chocolate square falls into frame from above and lands at the center with a soft bounce. On impact, chocolate shavings, cocoa powder and gold foil flakes ripple outward in rings like a shockwave, self-arranging into a perfectly symmetrical eight-petal rangoli in about 1.5 seconds. Movement is fast at impact then eases out as the pattern settles. Gold flakes catch the light with small glints. End on the completed rangoli exactly as in Image B, perfectly still. Camera static, focus constant. Aspect ratio 9:16.
Slow-motion, high-frame-rate feel. Starting composition: Image C, the chocolate stack with the wrapped bar beside it. A thick ribbon of molten milk chocolate enters from the top of frame and folds lazily over the top bar, splitting into smaller rivulets that run down the stepped edges with realistic viscous physics. Camera performs a slow push-in and slight arc to the right. The rim light shimmers across the moving chocolate surface. The wrapped bar stays perfectly still and the "digitalrushi" logo stays legible. End on a tighter framing with chocolate pooling at the base. Aspect ratio 9:16.
Start: Image D, the woman holding the bar over one eye, smiling. She lowers the bar slightly to reveal her eye, gives a playful wink, then laughs and tilts her head. Hair moves naturally. Camera does a gentle handheld-feel drift left with a subtle push-in. Background gold bokeh twinkles softly. The bar keeps its shape and logo through the motion and her hand anatomy stays correct, five fingers throughout. End on a warm laughing expression with the bar held beside her cheek. Aspect ratio 9:16.
Extreme macro, slow-motion. Starting composition: Image E. Fingers apply pressure and the chocolate square snaps cleanly away from the bar; a few crumbs burst outward and drift down slowly. Camera static with a micro rack-focus from the bar edge to the separated square. Hard side light stays consistent. End with the square held slightly apart from the bar, crumbs settling. Aspect ratio 9:16.
Start with the gift box from Image F held by the giver's hand on the left. It slides across the table and the receiver's hand enters from the right to take it. Diya flames flicker gently in the foreground blur. Camera performs a slow lateral dolly following the box left to right. Focus stays on the box, and the "digitalrushi" logo stays legible. End with both hands holding the box together at center frame. Natural hand anatomy throughout. Aspect ratio 9:16.
Start: exactly Image G — wide frontal night shot of the city junction with the vertical purple billboard centered. Camera performs one smooth, continuous, straight-line push-in directly toward the center of the billboard. No rotation, no tilt, no sideways drift; the billboard stays perfectly centered and frontal the entire time. Speed: slow for the first second, then a steady accelerating dolly. The street, traffic light trails and fairy lights move past the frame edges and fall out of view as the billboard grows. The billboard keeps its exact rectangular shape and straight edges throughout and does not warp or bend. The screen content stays flat on the screen surface. The final 1 second: the billboard's edges pass beyond the frame borders and the screen fills the whole frame, faint LED pixel grid visible, landing exactly on Image H. Lighting stays consistent: purple screen glow, warm gold street lights. Aspect ratio 9:16.
Tip: the rangoli bloom in Shot 1 is the hardest motion to get — AI video tools handle things breaking apart much better than things assembling. If it won't come together, generate it in reverse (start from the finished rangoli and prompt it to scatter and collapse into the square), then reverse the clip in your editor. For the billboard, the screen content in Shot 6 is only a placeholder: cut Shots 1–5 into a fast ~4-second recap (about 0.5s per cut, ending on the rangoli), then corner-pin or planar-track it onto the billboard screen in After Effects or DaVinci Resolve, adding a light LED pixel-grid overlay so it looks emitted rather than pasted on. Because the last frame matches Shot 1, the reel loops cleanly. If you'd rather skip compositing, keep the static packshot on the billboard and just do the push-in.
Try It Yourself
The exact 13-image + ten shot-by-shot video prompts behind our masala potato chips reel — one potato drops from a giant hopper, gets washed, rolls down an amber-lit steel corridor, is wire-cut into slices, fried in hot oil, caught in a masala storm, and lands in a bag on a city street. Copy any card and paste it into your own tools.
The rule that holds this whole reel together: the hero object (potato, then slice, then chip) stays centred in frame and is always exactly ONE object — it never duplicates, splits or changes size. Every prompt says this for exactly that reason. Keep the chip thin and cupped from the fryer onward; generate Image E2 first and use it as the reference for F, G, H1 and H2 so the chip shape matches.
Commercial packshot of a premium potato chips pillow-pack bag standing upright, matte black film with subtle soft-touch texture, an orange circular swirl icon in the upper third, beneath it the brand name "digitalrushi" in lowercase bold rounded sans-serif with "digital" in white and "rushi" in orange (#E8692A), below that small white text "POTATO CHIPS" and bold orange text "SPICY MASALA", a photograph of red-orange masala-dusted chips and dried red chillies across the bottom of the bag, crimped heat-sealed top and bottom edges with fine serrations, bag gently puffed with air, standing on a brushed dark-steel surface, dark charcoal background, single warm horizontal LED strip light above casting a soft amber highlight across the bag's top edge, 85mm lens, f/5.6, sharp focus across the whole bag, photorealistic premium snack advertising photography, clean legible typography, single product, no extra text, 9:16
Photorealistic extreme low-angle shot looking straight up at a giant industrial stainless-steel funnel hopper filling the upper two-thirds of the frame, packed to the brim with hundreds of clean whole russet potatoes visible over its rim and through a narrow glass inspection window on its side, brushed steel with rivets and weld seams, a square trapdoor at the hopper's base just opening, a single potato beginning to drop out of it, sleek dark industrial interior around it, warm amber LED strip lines running along the hopper's edges and the ceiling, slight haze in the air catching the light, 16mm wide-angle lens for dramatic scale, deep focus, symmetrical composition centered on the trapdoor, photorealistic premium FMCG commercial aesthetic, 9:16
Photorealistic front-on shot of a single whole russet potato rolling along a brushed dark-steel chute toward camera, passing through a thin, glassy curtain of falling clear water, droplets bursting outward and frozen mid-air, potato skin glistening wet and clean, symmetrical steel channel walls on both sides, warm amber LED strip lines along the walls reflecting in the wet steel floor, dark industrial background, 35mm lens at chute level, f/4, fast shutter look freezing every droplet, photorealistic premium FMCG commercial aesthetic, 9:16
Photorealistic one-point-perspective shot down a long, perfectly symmetrical brushed-steel tunnel, warm amber LED strip lines running along the upper edges of both walls and converging at a distant vanishing point, two polished steel guide rails along the floor, a single clean whole russet potato rolling along the rails in the center of frame toward camera, soft reflections on the polished steel floor, faint haze in the distance, camera at rail level, 35mm lens, deep focus, minimal and premium, photorealistic FMCG commercial aesthetic, 9:16
Photorealistic front view at roller level of a single clean whole russet potato resting on a bed of polished horizontal stainless-steel rollers, directly beneath a heavy brushed-steel slicing frame, the frame's top bar spanning the full width of the shot with a row of tightly spaced thin vertical stainless-steel wires hanging down from it, the wires lowered so they pass in front of and through the potato, cool white light glinting along each wire, the potato appearing completely intact, dark brushed-steel machine housing below the rollers with a round polished steel bolt detail at center, warm amber LED strip light along the back wall, 50mm lens, f/5.6, symmetrical centered composition, photorealistic premium FMCG commercial aesthetic, 9:16
Photorealistic front view at roller level of the same potato on the polished steel rollers, the slicing frame now lifted away out of the top of frame, the potato revealed as cut into thirty thin, perfectly even, round vertical slices still pressed together in the potato's original oval shape like a sliced loaf, fine straight cut lines visible across the brown skin, the leftmost slices just beginning to peel away and tip outward, pale cream raw potato flesh visible in the gaps, warm amber LED strip light along the back wall, dark brushed-steel housing with a round steel bolt below, 50mm lens, f/5.6, centered composition, photorealistic premium FMCG commercial aesthetic, 9:16
Photorealistic centered front-on shot of a single thin, round raw potato slice floating upright in mid-air facing camera, pale cream and slightly translucent with a faint star-shaped vein pattern at its center and a thin rim of brown skin, hovering just above a tall glass tank of shimmering hot golden cooking oil set inside a dark brushed-steel chamber, warm amber light glowing up through the oil, crossing amber light lines on the steel walls behind, faint heat shimmer, dark steel housing with a round polished bolt below the tank, 85mm lens, f/4, photorealistic premium snack commercial aesthetic, 9:16
Photorealistic centered front-on shot of a single thin, crisp golden potato chip, about 2 mm thick, floating upright just above the surface of the same glass tank of hot golden oil, its sides curled forward into a gentle cupped shape with a rounded bottom edge, surface covered in fine oil blisters, a few droplets of golden oil dripping from its lower edge back into the tank, steam wisps rising, the oil surface still bubbling below, dark brushed-steel chamber with amber light lines behind, 85mm lens, f/4, only one chip in frame, photorealistic premium snack commercial aesthetic, 9:16
Photorealistic centered front-on shot of the same thin golden cupped potato chip (Image E2), about 2 mm thick, floating upright in mid-air inside a dark charcoal brushed-steel chamber above polished steel rollers, not touching them, a dense column of red-orange masala spice powder (#D9481E) pouring down onto it from a polished steel nozzle at the top of frame and bursting outward around its thin curled edges in a swirling cloud, fine chilli and spice flecks glinting in warm amber light, the chip's top half already dusted red-orange, 85mm lens, f/4, fast shutter look freezing the particles, the chip stays thin and cupped, never puffy or thick, photorealistic premium snack commercial aesthetic, 9:16
Photorealistic shot of the thin red-orange masala-seasoned cupped potato chip mid-tumble, tilted at an angle, flying away from camera down a long symmetrical brushed-steel tunnel, warm amber LED strip lines along both upper edges converging toward a brilliant white daylight opening at the far end, strong backlight from the opening rim-lighting the chip's curled edges, polished floor reflecting the light, slight motion blur on the tunnel walls, chip in sharp focus, 35mm lens, deep focus, photorealistic premium snack commercial aesthetic, 9:16
Photorealistic bright daylight shot of an Indian city street in front of a row of closed silver corrugated steel rolling shutters, a young Indian man in his mid-20s in a beige overshirt over a navy t-shirt and dark jeans walking past from left to right in mid-stride, holding the open matte black "digitalrushi" Spicy Masala chips bag (Product Anchor) at waist height with the logo facing camera, a single thin red-orange seasoned potato chip arcing through the air above him toward the open bag, crisp sunlight, clean minimal composition, 35mm lens, f/4, chip and bag in sharp focus, natural hand anatomy, photorealistic premium snack commercial aesthetic, 9:16
Photorealistic close-up of the open matte black "digitalrushi" Spicy Masala potato chips bag (Product Anchor) held up by one hand at the bottom of frame, the bag's open top overflowing with thin golden red-orange masala-dusted chips, a few chips and spice flecks flying upward above it, background of softly blurred silver steel rolling shutters in bright daylight, logo, "POTATO CHIPS" and "SPICY MASALA" clearly legible, top 30% clean light space for text, 50mm lens, f/2.8, natural hand anatomy with five fingers, photorealistic premium snack advertising photography, 9:16
Generate each clip from its matching reference image, then cut them together in order. Keep each stage to 2–3 seconds in the edit — the reel loses its pace if any single stage runs long.
Starting composition: exactly Image A — extreme low angle under the giant potato-filled steel hopper, perfectly centered and symmetrical. The trapdoor at its base snaps open with a sharp mechanical motion; potatoes above shift slightly, and a single potato drops out alone and falls straight down toward camera with realistic weight and a slight tumble. Only ONE potato falls; the trapdoor closes immediately; no other potatoes leave the hopper. Camera holds steady for the first second, then tilts down fast to follow the potato as it falls past the lens. Lighting: warm amber LED lines constant, faint haze. End with the potato falling out of the bottom of frame. Aspect ratio 9:16.
Starting composition: the potato lands on the steel chute with a small bounce, the water curtain falling steadily ahead of it. It rolls forward toward camera, punches through the water curtain, droplets spraying outward in slow motion, and emerges glistening and clean. Camera at chute level slowly pulls back as the potato advances, keeping it centered. Amber reflections slide across the wet steel. Speed: real time, easing to slow motion through the curtain, back to real time after. End with the wet potato centered, rolling toward camera. Aspect ratio 9:16.
Starting composition: one-point perspective down the amber-lit steel tunnel, potato small in the distance on the rails. The potato rolls steadily toward camera along the guide rails, growing in frame while staying perfectly centered. Camera performs a slow backward dolly so the potato gains on it gently. Amber LED lines stay fixed and symmetrical; reflections move naturally across the polished floor. No camera shake. End with the potato large in the center of frame, just reaching a set of vertical steel wires ahead. Aspect ratio 9:16.
Starting composition: exactly Image D1 — the whole potato resting on polished steel rollers beneath the heavy steel slicing frame, vertical wires hanging down. The frame drives straight down through the potato in one fast, clean, mechanical stroke; the wires pass completely through it and the potato does not move, crumble or deform — it still looks perfectly whole. A brief half-second hold. Then the steel frame lifts straight up and slides away out of the top of frame, revealing that the potato has been cut into thirty thin, even, round slices that remain pressed together in the exact oval shape of the potato, fine cut lines visible across the skin. Camera: locked off, perfectly centered at roller level, with a very slow push-in. Lighting: warm amber strip constant, cool glints on the wires. Physics: precise, weighty, industrial; no juice spray, no crumbs, no slices falling yet. End on exactly Image D2. Aspect ratio 9:16.
Starting composition: the potato cut into thin round slices still pressed together in its oval shape on the polished steel rollers. Starting from the left end, the slices peel away one at a time in a smooth sequence, like a deck of cards fanning out: each slice tips outward, separates from the next, and settles standing upright on its edge on the rollers with a small wobble, leaving an even gap between neighbours. Every slice is a pale cream, round raw potato cross-section with a thin rim of brown skin and a faint star-shaped vein pattern at its center. Slices never merge, duplicate, bend or disappear. As they fan out, the camera tracks smoothly to the right along the rollers at the same speed. When the fanning reaches the middle, a single round slice stays standing in the center of frame while the others drift out of view; it then lifts gently off the rollers and floats upward into the air, rotating very slightly to face camera. Camera keeps tracking with the floating slice as it drifts toward the glowing fryer chamber. End with the single raw slice floating centered. Aspect ratio 9:16.
Starting composition: exactly Image E1 — a single thin round raw potato slice floating upright and centered just above a tall glass tank of shimmering hot golden oil, inside the dark steel chamber with warm amber light lines. The slice lowers slowly straight down into the oil. The moment it touches, the oil erupts in dense tiny bubbles and wisps of steam rise around it. Fully submerged, it stays centered and upright as it cooks: the pale flesh turns opaque, then golden-yellow, fine oil blisters form across its surface, and its sides curl forward into a cupped chip shape. It then rises straight back up out of the oil, golden, with a few oil droplets dripping from its lower edge. There is only ONE slice and ONE chip in the frame at all times: the slice itself becomes the chip; no second chip appears, nothing else is in the oil. The chip never splits, duplicates, stacks or changes size. Camera: locked off, centered, slow push-in. End on exactly Image E2. Aspect ratio 9:16.
Starting composition: a single thin, crisp golden potato chip, about 2 mm thick, with its sides curled forward into a cupped shape and a rounded bottom edge, floating upright in mid-air at the center of the dark steel chamber, above the steel rollers, not touching them. A narrow column of red-orange masala powder pours straight down from a steel nozzle at the top of frame onto the chip, then bursts outward around its thin curled edges into a swirling cloud, particles glinting in the amber light. The spice settles evenly onto the chip's surface, turning it red-orange. The cloud drifts away, revealing the seasoned chip. The chip stays thin, cupped, upright, centered and the same size throughout; it never thickens, puffs up, swells or becomes a solid block, and it never lands on the rollers. Camera: locked off, centered, slow push-in. End with the thin red-orange seasoned chip floating still, faint spice haze behind it. Aspect ratio 9:16.
Starting composition: the seasoned chip floating centered in the dark chamber, facing camera. The chip tips forward and launches away from camera, tumbling end over end in a smooth, fast arc down a long brushed-steel tunnel toward a brilliant white daylight opening at the far end. The camera chases right behind it at the same speed, keeping the chip centered, warm amber light lines streaking past on both walls. The light grows brighter as the opening approaches, until the frame is flooded with white daylight as the chip flies out. Motion intensity: fast and energetic, contrasting with the calm previous shots. End on a near-white frame with the chip at the opening. Aspect ratio 9:16.
Starting composition: a near-white frame of daylight, resolving quickly into a bright Indian street in front of silver steel rolling shutters. The seasoned chip flies in from the top left in a smooth arc across the frame. A young man walks briskly from left to right, holding the open "digitalrushi" bag at waist height without breaking stride; the chip drops neatly into the open top of the bag and lands on the pile with a small bounce and a tiny puff of red spice. Camera: smooth tracking pan right, following the chip and then the bag, ending in a medium shot of the bag in his hand. Logo stays legible and unwarped, hands natural. End with the bag held steady at center frame. Aspect ratio 9:16.
Starting composition: the open bag held up close to camera, chips piled inside. The hand lifts the bag slightly toward camera; a few chips bounce up out of the top and fall back, and red spice flecks drift down in the sunlight. Camera: slow push-in, centered, focus on the chips and logo. The background shutters stay softly blurred. Logo, "POTATO CHIPS" and "SPICY MASALA" stay sharp and legible. End with a still, clean packshot leaving the top of the frame empty for the headline "WASHED. SLICED. SPICED." (added in edit). Aspect ratio 9:16.
Tip: the two shots most likely to go wrong are the fryer and the masala storm. In the fryer, watch for a second chip appearing in the oil — use E1 and E2 as start and end frames and keep the "only ONE slice and ONE chip" line in the prompt. In the masala storm, watch for the chip turning thick and puffy — upload Image E2 as the chip reference so the shape stays thin and cupped. If the hopper drops more than one potato, regenerate with the "only ONE potato falls" line kept in. Close the reel by cutting from the final packshot straight back to the hopper trapdoor opening, so it loops cleanly. And remove any tool watermarks by zooming each clip in about 5%.
Try It Yourself
The exact 7-image + 8 short video prompts behind our “tap the phone, the drink builds itself” reel — a phone lies upside down on a wine glass like a lid, and every single fingertip tap drops one ingredient into the glass in real recipe order: ice, juice, a stir, then a watermelon garnish and a 360° finish. Copy any card and paste it into your own tools.
The one rule that makes this mechanic work: every ingredient exists only inside the glass, and the top surface of the phone stays completely bare. Early tests had ice cubes “dancing” on top of the phone — adding explicit lines about where an ingredient can never appear fixed it. Every video prompt below states this on purpose.
Hyper-realistic cinematic photograph, vertical 9:16, a front-on shot at eye level of a clear stemmed wine glass with a gold-tinted stem standing in the lower centre of a dark green marble bar counter with soft veins and a polished reflective surface. The background is pure black with a faint warm bokeh of golden bottles. A woman's hand enters from the top-left holding a large plain black smartphone in landscape orientation close to the camera, tilted slightly towards the viewer, so the phone fills about half the frame width and sits in the upper half of the image above the glass. The phone screen faces the camera, fully lit and sharply in focus, showing a bright, high-resolution, saturated photo of a halved watermelon with deep red flesh, black seeds, a fresh slice and green mint leaves on a clean white background, with every detail crisp and clearly readable. A subtle glow from the screen lights the hand and the top of the glass. The hand is natural with five fingers and neat nails, gripping the phone by its edges without covering the screen. Dramatic low-key lighting: a warm key light from the right creating a bright rim on the glass, deep shadows around. 50mm lens, deep enough depth of field to keep both the phone screen and the glass sharp, crisp reflections on the marble. No logos on the phone, no text.
Hyper-realistic cinematic photograph, vertical 9:16, a front-on shot at eye level of a clear stemmed wine glass with a gold-tinted stem standing in the lower centre of a dark green marble bar counter with soft veins and a polished reflective surface. The background is pure black with a faint warm bokeh of golden bottles. A large plain black smartphone with no logo lies upside down across the rim of the glass like a lid, its back facing up and seen from the front as a thin dark slab. A woman's hand enters from the top-left, with a single index finger pressing lightly on the back of the phone from above and the other fingers relaxed; the hand is natural with five fingers and neat nails. Clear ice cubes fall in mid-air from the underside of the phone straight down into the glass, some still tumbling, with the glass already filled with glassy ice cubes up to about three-quarters, tiny frost droplets on the glass and a bright rim highlight. No liquid yet. Dramatic low-key lighting: a warm key light from the right creating a bright rim on the glass, deep shadows around. 50mm lens, deep enough depth of field to keep both the phone and the glass sharp, crisp reflections on the marble. No logos, no text.
Hyper-realistic cinematic photograph, vertical 9:16, a front-on shot at eye level of a clear stemmed wine glass with a gold-tinted stem standing in the lower centre of a dark green marble bar counter with soft veins and a polished reflective surface. The background is pure black with a faint warm bokeh of golden bottles. A large plain black smartphone with no logo lies upside down across the rim of the glass like a lid, its back facing up and seen from the front as a thin dark slab. A woman's hand enters from the top-left, with a single index finger pressing lightly on the back of the phone from above and the other fingers relaxed; the hand is natural with five fingers and neat nails. A thin stream of vibrant watermelon-pink juice falls from the underside of the phone straight down into the glass. The glass is filled with glassy ice cubes and watermelon-pink juice nearly to the top, glowing from within in the warm light, with fine bubbles and a soft pink-to-deeper-red gradient toward the bottom, and condensation droplets covering the glass. Dramatic low-key lighting: a warm key light from the right creating a bright rim on the glass, deep shadows around. 50mm lens, deep enough depth of field to keep both the phone and the glass sharp, crisp reflections on the marble. No logos, no text.
Hyper-realistic cinematic photograph, vertical 9:16, a front-on shot at eye level of a clear stemmed wine glass with a gold-tinted stem standing in the lower centre of a dark green marble bar counter with soft veins and a polished reflective surface. The background is pure black with a faint warm bokeh of golden bottles. A thin black smartphone with no logo lies upside down flat across the rim of the glass like a lid, its back facing up and seen from the front as a thin dark slab, with a bare, clean top surface. A woman's hand enters from the left with only the tip of the index finger extended, hovering about one centimetre above the phone. The glass is filled with glassy ice cubes and vibrant watermelon-pink juice up to just below the rim, with a clear bright pink at the top and a deeper red gradient at the bottom, fine bubbles and condensation droplets on the glass. The juice is still, with no stream falling. Dramatic low-key lighting: a warm key light from the right creating a bright rim on the glass, deep shadows around. 50mm lens, deep enough depth of field to keep both the phone and the glass sharp, crisp reflections on the marble. No logos, no text.
Hyper-realistic cinematic photograph, vertical 9:16, the same front-on shot at eye level, same clear stemmed wine glass with a gold-tinted stem, same dark green marble bar, same black background with warm golden bokeh and same lighting as the previous image. The thin black smartphone lies upside down flat across the rim of the glass like a lid, its back facing up. A long slim polished steel bar stirrer with a twisted shaft stands upright in the centre of the glass, passing through the phone like a stirrer through a cocktail lid: its lower half is submerged in the juice inside the glass, and its upper end rises about eight centimetres above the back of the phone, ending in a small round steel bead. A woman's hand enters from the left with only the tip of the index finger extended, hovering beside the stirrer's upper end and not touching it. The glass is filled with ice cubes and freshly stirred watermelon-pink juice up to just below the rim, the pink and red now blended into one smooth, glowing watermelon colour with a faint swirl and fine bubbles rising. Condensation droplets cover the glass. Dramatic low-key lighting: a warm key light from the right creating a bright rim on the glass and a sharp highlight along the steel stirrer, deep shadows around. 50mm lens, deep depth of field to keep the phone, the stirrer and the glass sharp, crisp reflections on the marble. No logos, no text.
Hyper-realistic cinematic photograph, vertical 9:16, the same front-on shot at eye level, same clear stemmed wine glass with a gold-tinted stem, same dark green marble bar, same black background with warm golden bokeh and same lighting as the previous image. The phone, the hand and the stirrer are gone. The glass stands alone with an open rim, filled with glassy ice cubes and smooth watermelon-pink juice up to just below the rim, with a soft swirl fading on the surface, a few tiny droplets on the inside of the glass above the juice, and condensation droplets running down the outside. A soft pink reflection lies on the marble. Dramatic low-key lighting: a warm key light from the right creating a bright rim on the glass, deep shadows around. 50mm lens, deep depth of field, crisp reflections on the marble. No hands, no phone, no logos, no text.
Hyper-realistic cinematic photograph, vertical 9:16, the same front-on shot at eye level, same clear stemmed wine glass with a gold-tinted stem, same dark green marble bar, same black background with warm golden bokeh and same lighting as the previous image. The glass is filled with glassy ice cubes and glowing watermelon-pink juice, with a half-moon slice of fresh watermelon sitting upright on the rim at the side of the glass, its small notch resting on the rim, showing deep red juicy flesh, tiny black seeds, a thin white pith and a green rind edge, with a single drop of juice running down its edge. Condensation droplets glisten on the glass and a soft pink reflection shimmers on the marble. Dramatic low-key lighting: a warm key light from the right creating a bright rim on the glass, deep shadows around. Premium beverage-commercial look, 85mm, shallow depth of field, crisp reflections. No hands, no phone, no logos, no text.
Use one fingertip of a single hand, one tap per ingredient, and a locked-off camera. Each card tells you which frame to start from.
Start frame: Image 1. About 4 seconds. Trim the result at roughly 3.0 seconds, where the phone is placed and the fingertip is hovering with no ice yet, and save that frame as the start of Video 1B.
A hand holds a black smartphone showing a watermelon photo, turns it upside down, and lays it flat on top of the wine glass like a lid. Static camera, one hand only.
Start frame: the frame at about 3.0 seconds of Video 1A (phone on the rim, fingertip hovering, no ice yet). About 6 seconds. Use a start frame only, no end frame.
Locked-off static camera at eye level, no camera movement at all. A clear stemmed wine glass with a gold-tinted stem stands on a dark green marble bar in front of a black background with warm golden bokeh. A thin black smartphone lies upside down flat across the rim of the glass like a lid, with the back of the phone facing up. The top surface of the phone stays completely bare, flat and motionless for the entire shot, with nothing on it or above it. A single hand enters from the left with only the tip of the index finger extended. At 0.5 seconds the fingertip taps the back of the phone once, softly, then rests about one centimetre above the phone and stays still for the remainder of the shot. After the tap, clear glassy ice cubes begin to fall one after another inside the glass only, appearing at the top of the glass's interior directly beneath the phone's underside and falling straight down through the empty glass under gravity. Each cube hits the bottom or the cubes below with a soft glassy clink, bounces slightly against the inner walls and settles. Over about four seconds roughly eight to ten cubes accumulate until the glass is three-quarters full, each cube catching sharp white highlights and refracting the warm bokeh lights. All cubes stay inside the glass, none appear on top of the phone, none float above the phone, none fall outside the glass, none land on the marble. The glass and the phone do not move or shake. Fine frost droplets form on the glass. Warm key light from the right with a bright rim on the glass, deep shadows, crisp reflections on the marble, shallow depth of field. Realistic physics and natural motion, no scene cuts. Only one hand and one fingertip are visible in the shot.
Start frame: the last frame of Video 1B. About 6 seconds. If it fails, upload Image 4 as the end frame.
IMPORTANT: the juice exists only inside the wine glass, below the phone. The top of the phone stays completely bare. Locked-off static camera at eye level, no camera movement at all. A clear stemmed wine glass with a gold-tinted stem stands on a dark green marble bar in front of a black background with warm golden bokeh, filled three-quarters with clear glassy ice cubes. A thin black smartphone lies upside down flat across the rim of the glass like a lid, with the back of the phone facing up. The top surface of the phone stays completely bare, flat and motionless for the entire shot, with nothing on it or above it. A single hand enters from the left with only the tip of the index finger extended. At 0.5 seconds the fingertip taps the back of the phone once, softly, then rests about one centimetre above the phone and stays still for the remainder of the shot. After the tap, a thin, smooth, slightly wavy stream of vibrant watermelon-pink juice emerges from the underside of the phone, inside the glass, directly beneath it, and falls straight down onto the ice under gravity. The juice splashes softly over the cubes, flows down between them and rises gradually over about four seconds from the bottom of the glass up to just below the rim, the colour glowing from within with a bright pink at the top and a deeper red gradient toward the bottom, fine bubbles rising, ice cubes shifting slightly and clinking as the liquid fills in. The juice stays entirely inside the glass: none appears on top of the phone, none floats above the phone, none spills over the rim, none lands on the marble. The glass and the phone do not move or shake. Condensation droplets form on the glass. Warm key light from the right with a bright rim on the glass, deep shadows, crisp reflections on the marble and a soft pink glow on the counter, shallow depth of field. Realistic liquid physics and natural motion, no scene cuts. Only one hand and one fingertip are visible in the shot.
Start frame: the last frame of Video 2 (or Image 4). About 5 seconds. Image 5 can be used as the end frame if needed.
IMPORTANT: the stirrer passes through the phone like a lid: its lower half is inside the glass and its upper end sticks out above the phone. The ice and juice stay inside the glass only. Locked-off static camera at eye level, no camera movement at all. A clear stemmed wine glass with a gold-tinted stem stands on a dark green marble bar in front of a black background with warm golden bokeh, filled with clear ice cubes and glowing watermelon-pink juice up to just below the rim. A thin black smartphone lies upside down flat across the rim of the glass like a lid, with the back of the phone facing up. Only the stirrer's top end rises above the phone. A single hand enters from the left with only the tip of the index finger extended. At 0.5 seconds the fingertip taps the back of the phone once, softly, then rests about one centimetre above the phone and stays still. After the tap, a long slim polished steel bar stirrer with a twisted shaft and a small round steel bead on top rises up from inside the glass, passing through the phone like a stirrer through a cocktail lid, until its top end stands about eight centimetres above the back of the phone. The lower half of the stirrer is in the juice. The stirrer then turns in two smooth circular turns around its own centre line while its lower half sweeps the juice, the ice cubes rotating gently and clinking against the glass, the juice swirling into a soft spiral as the pink and red layers blend into one smooth watermelon colour with fine bubbles rising. The upper end above the phone turns slowly in a small circle with it. The stirrer slows and rests still in the centre. Nothing else appears on top of the phone. The juice stays inside the glass, no liquid spills. The glass and phone do not move. Condensation droplets run down the glass. Warm key light from the right with a bright rim on the glass, deep shadows, crisp reflections on the marble, shallow depth of field. Realistic liquid physics and natural motion, no scene cuts. Only one hand and one fingertip are visible in the shot.
Start frame: the last frame of Video 3. About 4 seconds. Image 6 can be used as the end frame if needed.
Locked-off static camera at eye level, no camera movement at all. The same wine glass of ice and blended watermelon-pink juice stands on the dark green marble bar in front of the black background with warm golden bokeh, with the thin black smartphone lying upside down across its rim and the steel stirrer passing up through the phone with its top end standing above the back of the phone. A single hand enters from the left, the thumb and index finger gently pinching the edges of the phone. At 1 second the hand moves in and lifts the phone straight up off the rim. The steel stirrer stays attached to the phone, sliding up and out of the juice as the phone rises, a few tiny droplets falling from its lower end back into the glass. The hand carries the phone and the stirrer together up and out of the frame to the left. The glass stays still and the juice gently settles, with a soft swirl fading, the ice cubes drifting slightly, and a bright pink glow catching the rim light. Condensation droplets run slowly down the glass. Warm key light from the right, deep shadows, crisp reflections on the marble. Realistic physics, natural hand, no scene cuts. The shot ends on the glass standing alone, completely still, with the rim open.
Start frame: the last frame of Video 4. About 5 seconds. Image 7 can be used as the end frame if needed.
Locked-off static camera at eye level, no camera movement at all. The same wine glass of ice and blended watermelon-pink juice stands alone on the dark green marble bar in front of the black background with warm golden bokeh. A single hand enters from the right holding a half-moon slice of fresh watermelon between the thumb and fingertips, with deep red juicy flesh, tiny black seeds, a thin white pith and a green rind edge, with a small notch cut at the centre of the straight edge. At 1 second the hand gently slides the notch onto the rim of the glass at the side, so the half-moon slice sits upright on the rim, leaning slightly outwards with its flat edge on the glass. The hand lets go and moves out of the frame to the right. The slice stays in place and a tiny drop of juice runs slowly down its edge. The glass stays still, condensation droplets glisten, a soft pink reflection shimmers on the marble. The shot ends on a calm steady hold of the finished drink. Premium beverage-commercial look, warm key light from the right with a bright rim on the glass, shallow depth of field, crisp reflections. Realistic physics, natural hand with five fingers, no scene cuts.
Start frame: Image 7 (or the last frame of Video 5). About 8 seconds.
Start on the exact composition of the start image: the finished watermelon cooler in a clear stemmed wine glass with a half-moon watermelon slice on the rim, standing alone on the dark green marble bar in front of the black background with warm golden bokeh. The glass stays perfectly still. The camera performs a smooth, continuous, constant-speed orbit exactly 360 degrees around the glass at eye level, starting slowly with a gentle push-in of about 15% during the first second, then continuing the orbit at a steady pace, circling the glass fully and returning to the starting view by the end of the shot. During the orbit, the half-moon slice on the rim passes through the front, the side and the back views, the glowing watermelon-pink juice, the ice cubes and the condensation droplets catch the warm rim light from different angles, and a soft pink reflection slides across the marble. The background stays black with warm golden bokeh, consistent from every angle. Smooth gimbal motion with no shake and no speed changes. Realistic physics, consistent lighting, no scene cuts. No hands, no phone, no text.
Use this instead of the camera orbit if the background or glass wobbles during the orbit. Start frame: Image 7. About 8 seconds.
Start on the exact composition of the start image. The camera is locked static at eye level with a slow push-in of about 15% over the first second. The glass stands on a polished dark green marble round turntable that rotates smoothly and continuously at constant speed, turning the glass a full 360 degrees in 7 seconds, so the half-moon watermelon slice on the rim travels through the front, side and back views. The black background with warm golden bokeh stays fixed. The juice, the ice cubes and the condensation droplets catch the warm rim light from every angle, a soft pink reflection slides across the marble. Smooth motion with no shake and no speed changes, realistic physics, no scene cuts. No hands, no phone, no text.
Start frame: the last frame of Video 6A. About 6 seconds.
Start on the finished watermelon cooler from the same eye-level front view. The camera performs a smooth continuous push-in to a tight macro close-up in the first 2 seconds, moving towards the upper half of the glass, where a drop of juice runs down the half-moon watermelon slice on the rim and tiny condensation droplets glisten on the glass. The camera then glides in a slow arc of about 45 degrees to the right while holding the close-up, passing the glassy ice cubes and the glowing pink juice with fine bubbles rising inside. The camera finally drifts gently down to the base of the glass, showing the golden stem and the pink reflection on the marble. The glass stays perfectly still. The background stays black with warm golden bokeh. Shallow depth of field with sharp focus on the nearest surface, crisp macro detail, warm key light from the right with bright rim highlights. Smooth gimbal motion, no shake, realistic physics, no scene cuts. No hands, no phone, no text.
Tip: the three things most likely to go wrong are (1) ice or juice appearing on top of the phone — repeat the “IMPORTANT” line at the start of the prompt and add “the phone is a solid opaque slab, items emerge only from beneath it”; (2) a second hand appearing — keep “only one hand and one fingertip are visible” at the end of the prompt; and (3) the glass or background changing between clips — always start each clip from the previous clip’s last frame and never reword the scene description. If a clip still fails, shorten it to 4 seconds or upload the next image in the list as the end frame. For the final 360° shot, use the turntable alternative if the background wobbles during the camera orbit. Add a short hook (“Tap. Mix. Sip.”) and a call to action in your editor rather than in the prompt, because AI tools often misspell on-screen text.
Try It Yourself
The exact 5-image + short video prompts behind our “helicopter unveils a giant Diwali sweet box” reel — a giant maroon-and-gold mithai box rests on a lit-up festive street under a gold cloth, a helicopter flies in with its cable attached, hooks the cloth and lifts it away, and the lid opens in a burst of golden light, sweets and fireworks. Copy any card and paste it into your own tools.
The one rule that makes this mechanic work: the helicopter and its cable are one connected object — the cable must hang from the helicopter’s belly in every frame and never drift away from it. Early tests had the helicopter on one side of the frame and the rope on the other. Every video prompt below states the attachment on purpose, and the box always rests firmly on the ground.
Hyper-realistic cinematic photograph, vertical 9:16, a lively Indian Diwali market street at dusk with a deep blue-orange sky, shop fronts with strings of warm fairy lights, marigold garlands across the street, glowing brass diyas and colourful rangoli on the road, and lit-up balconies. In the centre of the street, a giant tall rectangular object about fifteen metres tall stands firmly on the road, covered from top to bottom by a loose glossy gold satin cloth with deep natural folds, gathered into a peak at the top and draping down to the ground, where the hem pools around a rangoli. In the upper right of the frame, a white helicopter flies at a distance, small in the sky, with its rotor blades spinning. A single thin steel cable hangs from the centre of the helicopter's belly, about twenty metres long, with a steel hook at its lower end swinging slightly behind it. The cable is clearly attached to the helicopter. A few tiny people stand far back at the edges of the street for scale. Low-angle wide 24mm, sharp detail, rich saturated warm colours, glowing bokeh. No text, no logos.
Hyper-realistic cinematic photograph, vertical 9:16, the same Diwali market street, lights, rangoli and dusk sky as the previous image. The giant tall rectangular object still stands firmly on the road, covered by the glossy gold satin cloth, its hem on the ground around the rangoli and its soft shadow on the street. A white helicopter hovers directly above it at the top of the frame, rotor blades spinning, with a single thin steel cable running vertically straight down from the centre of its belly to the peak of the gold cloth, the steel hook caught in the peak, which is pulled upward into a tight tall point with deep folds radiating down the sides. The helicopter, the cable and the peak are aligned on one vertical line. Fairy lights and marigold garlands glow in the background. Low-angle wide 24mm, sharp detail, rich saturated warm colours, glowing bokeh. No text, no logos.
Hyper-realistic cinematic photograph, vertical 9:16, the same Diwali market street, lights and dusk sky. The white helicopter hovers higher above the box, lifting the glossy gold satin cloth upward on a single thin steel cable attached to the centre of its belly, the cloth billowing into a golden canopy that has peeled away from the top half of a giant maroon gift box. The revealed top half shows a glossy maroon lid with ornate gold borders, gold paisley patterns and a gold ribbon-style band, and the top of the front panel. The lower half of the box is still covered by the gold cloth, its hem on the road. The helicopter, cable and the lifted peak of the cloth are aligned on one vertical line. Warm sparkles in the air and glowing fairy lights all around. Low-angle wide 24mm, sharp detail, rich saturated colours. No text, no logos.
Hyper-realistic cinematic photograph, vertical 9:16, a giant tall rectangular mithai gift box about fifteen metres tall standing firmly on the road in the middle of a lively Indian Diwali market street at dusk, photographed straight-on from a low angle with its front fully facing the camera. The box is deep maroon with ornate gold borders, gold paisley and diya patterns, a gold ribbon-style band and a closed glossy maroon lid. On the front panel, the words "DIGITALRUSHI SWEETS" are printed in large bold gold serif capital letters centred horizontally, with a thin gold line beneath them and the smaller gold text "HAPPY DIWALI" below, all crisp, perfectly spelled, evenly spaced and undistorted. Marigold garlands and glowing brass diyas surround the base, with a colourful rangoli on the road and a soft contact shadow under the box. Strings of warm fairy lights across the street, a deep blue-orange dusk sky. A white helicopter flies off in the upper right with the gold satin cloth hanging from a single steel cable attached to its belly, the cloth trailing in the wind. A few tiny people at the edges of the street show the scale. Wide-angle 24mm, sharp detail, rich saturated warm colours. No other text, no other logos.
Hyper-realistic cinematic photograph, vertical 9:16, the same giant maroon-and-gold mithai gift box on the same Diwali market street at dusk, photographed straight-on from a low angle, now with its lid hinged fully open at the top, a column of warm golden light beaming upward from inside, and golden fireworks bursting in the dusk sky above. Inside, a glowing heap of colourful Indian sweets is visible: golden laddoos, white-silver kaju katli diamonds, orange jalebi rings and pink barfi, with fine sparkles floating up. On the front panel, the words "DIGITALRUSHI SWEETS" in large bold gold serif capital letters with the smaller gold text "HAPPY DIWALI" below, crisp, perfectly spelled and undistorted. Marigold petals drift down through the glowing air, diyas and rangoli surround the base. Wide-angle 24mm, low angle, sharp detail, rich saturated warm colours. No other text, no other logos.
Use the start frame only (no end frame) for these short prompts. Each card tells you which frame to start from.
Start frame: Image 1. About 8 seconds. If the cable and helicopter drift apart, use the two short clips (1A and 1B) below instead.
The first frame is exactly the start image. The camera is a wide static shot from street level looking up, with a very slow push-in of about 3%. The white helicopter flies in from the upper right in a smooth curved path across the dusk sky towards the box, its rotor blades spinning. A single thin steel cable is attached to the centre of its belly for the entire shot and hangs below it, trailing slightly behind and swaying gently as the helicopter travels, with the steel hook swinging at its lower end. The cable and hook always remain connected to the helicopter and travel with it; they never separate or move on their own. From 0 to 4 seconds, the helicopter travels steadily across the sky, growing larger as it approaches. At about 4 seconds it slows and moves into position directly above the peak of the gold cloth, the cable straightening until it hangs perfectly vertically beneath the helicopter. From 5 seconds the helicopter lowers slightly while the cable lengthens, bringing the hook down along the same vertical line towards the peak of the cloth. At about 7 seconds the hook catches the peak of the cloth and the cable tightens, pulling the peak up into a tall point with deep folds radiating down the sides. The helicopter, the cable, the hook and the peak are aligned on one vertical line. The box stays firmly on the ground, the gold cloth ripples from the rotor wash, fairy lights twinkle. Realistic cable physics, consistent dusk light, no scene cuts. The last frame matches Image 2.
Start frame: Image 1. About 4 seconds.
The first frame is exactly the start image. Static wide camera at street level looking up, with a very slow push-in of about 3%. The white helicopter flies in from the upper right in a smooth curve towards the box and slows to a hover directly above the peak of the gold cloth. A single thin steel cable is attached to the centre of its belly for the entire shot, trailing behind and swaying slightly as it flies, then straightening to hang vertically as the helicopter stops. The cable and hook never separate from the helicopter. No scene cuts.
Start frame: the last frame of Video 1A. About 4 seconds.
Static wide camera at street level looking up. The white helicopter hovers steadily directly above the covered box, its position fixed. The single steel cable stays attached to the centre of its belly and hangs vertically straight down. The cable lengthens slowly, lowering the hook along the same vertical line towards the peak of the gold cloth. At 3 seconds the hook catches the peak and the cable tightens, pulling the peak up into a tall point with deep folds. The helicopter, cable, hook and peak stay aligned on one vertical line. The box stays firmly on the ground. No scene cuts.
Start frame: Image 2 (or the last frame of Video 1). About 5 seconds.
The first frame is exactly the start image. The camera is a wide static shot from a low angle with a very slow push-in of about 5%. The cable stays attached to the centre of the helicopter's belly and to the hook in the cloth's peak at all times, with the helicopter always directly above the cable and the cloth, aligned on one vertical line. The helicopter rises slowly and steadily, pulling the gold satin cloth upward on its cable, peeling it away from the top of the box and billowing into a golden canopy, revealing the glossy maroon lid with its gold borders, paisley patterns and ribbon band, with sparkles in the air. The box stays firmly on the ground and its lower half stays covered. Realistic cloth physics, realistic rotor wash, consistent dusk light, no scene cuts. The last frame matches Image 3.
Start frame: Image 3 (or the last frame of Video 2). About 5 seconds.
The first frame is exactly the start image. The camera is a wide static shot from a low angle with a very slow push-in of about 6%. The cable stays attached to the centre of the helicopter's belly and to the cloth at all times, with the helicopter directly above the cable and the cloth. The helicopter lifts and then slides to the upper right, pulling the gold satin cloth completely off the box in one smooth sweep, the cloth billowing and rippling in the rotor wash and trailing from the cable behind the helicopter. The full maroon-and-gold box is revealed, with its front panel text "DIGITALRUSHI SWEETS" and "HAPPY DIWALI", glowing fairy lights and diyas around its base. The helicopter flies off to the upper right. The box stays firmly on the ground, and the text stays sharp, correctly spelled and undistorted. Realistic cloth physics, consistent dusk light, no scene cuts. The last frame matches Image 4.
Start frame: Image 4 (or the last frame of Video 3). About 5 seconds.
The first frame is exactly the start image. The camera is a wide static shot from a low angle with a smooth push-in of about 10%. At 1 second the glossy maroon lid hinges open slowly at the back edge like a gift box, a column of warm golden light beaming upward from inside, and golden fireworks burst in the dusk sky above, lighting the street. Glowing sweets heap up inside the box and fine sparkles float into the air, marigold petals drift down. The helicopter exits the frame. The text "DIGITALRUSHI SWEETS" and "HAPPY DIWALI" on the front stays flat, sharp and undistorted. The box stays firmly on the ground. Realistic light and particle motion, consistent dusk light, no scene cuts. The last frame matches Image 5.
Tip: the three things most likely to go wrong are (1) the helicopter and cable separating — add “the cable is a rigid part of the helicopter, hanging from its belly, never apart from it” and keep the cable length the same in every prompt; (2) the box floating instead of resting — add “the base of the box rests firmly on the road with a contact shadow and the rangoli visible around its base”; and (3) garbled text on the box — repeat “the word DIGITALRUSHI, spelled D-I-G-I-T-A-L-R-U-S-H-I”, or generate the front panel blank and add the text in your editor. If the single 8-second flight-and-anchor clip fails, use the two short clips (1A and 1B). Add a hook (“Diwali sweets, launched like a Hollywood movie”) and a call to action in your editor rather than in the prompt, because AI tools often misspell on-screen text.
Try It Yourself
The exact 5-image + 5 short video prompts behind our “Diwali sweet train” reel — a glass hyperloop train packed with DigitalRushi mithai races over a river town at dusk, slows down to show the sweets inside, pulls into a marigold-lit station and hands a hostel student far from home his ghar ki mithai. Copy any card and paste it into your own tools.
The one rule that makes this work: the train is always the same six glass carriages, every one packed with boxes. Every prompt below says “six carriages” and “the boxes never change” for exactly that reason — without it the train grows, shrinks or turns up empty between shots.
Hyper-realistic studio product photograph, square 1:1. A premium Indian festive sweets gift box: a square rigid box in deep maroon with an embossed gold-foil lattice (jaali) pattern and a thin burnt-orange border around the lid. The lowercase wordmark "digitalrushi" sits in the centre of the lid in gold foil, inside a slightly darker maroon panel, in a bold geometric sans-serif font, spelled exactly d-i-g-i-t-a-l-r-u-s-h-i. The lid leans open, and inside, on gold liners in neat compartments, are diamond-shaped kaju katli with edible silver leaf, orange motichoor laddoos, glossy gulab jamun in a small bowl and pistachio barfi. A lit brass diya, one marigold flower and a few loose petals sit beside the box. Warm key light from the left, soft gold rim light, deep maroon backdrop with soft bokeh fairy lights. Ultra-realistic textures: silver leaf shine, laddoo grains, syrup gloss. No other text, no watermarks.
Hyper-realistic cinematic aerial photograph, vertical 9:16, blue-hour dusk, a high drone angle looking down and across the scene (not straight down). A sleek six-carriage hyperloop train with a white aerodynamic bullet nose and a thin burnt-orange stripe along its full length runs on an elevated track that curves in a long S over a calm river and an old Indian town of flat rooftops. Every carriage has fully transparent glass walls and a glass roof, and inside every carriage are shelves packed with the maroon-and-gold digitalrushi boxes from the reference image and brass trays of laddoos and kaju katli, warmly lit. The train's nose is in the lower right of the frame and the carriages trail back along the curve into the distance. A continuous line of lit clay diyas runs along both edges of the track. On a rooftop in the lower left, a small family in festive clothes holds paper lanterns under marigold strings. Sky lanterns float up across a deep teal sky that reflects in the river. Warm amber against teal. No text, no watermarks.
Hyper-realistic cinematic aerial photograph, vertical 9:16, the same train, river, town, dusk light and camera direction as the reference image, but the camera is much closer to the train. The lead glass carriages now fill the right half of the frame, so the cargo is clearly visible through their transparent roof and walls: tiered shelves of maroon-and-gold digitalrushi boxes and round brass platters of motichoor laddoos, kaju katli and barfi, lit by warm golden interior light. The rest of the six-carriage train curves away into the distance over the river, the line of diyas glows along the track, and the rooftop family with lanterns is still in the lower left. No people inside the train, no text, no watermarks.
Hyper-realistic cinematic photograph, vertical 9:16, night. The same white train with its burnt-orange stripe has stopped at an elegant elevated station platform, seen from platform level at a three-quarter angle, with the carriages stretching back to the vanishing point. Large glass windows show lit shelves of maroon-and-gold digitalrushi boxes, trays of laddoos and kaju katli, and marigold garlands inside. The nearest glass door is closed. Thick marigold garlands and brass lanterns hang from the platform ceiling, a row of lit diyas runs along the platform edge, a large maroon, orange and gold rangoli covers the polished stone floor, and brass urlis filled with flowers and floating diyas stand in the foreground. Sky lanterns float in the teal sky beyond. No people, no text, no watermarks.
Hyper-realistic cinematic photograph, vertical 9:16, exactly the same platform, train, lighting and camera angle as the station reference image. Change only this: the nearest glass door has slid fully open, and in the doorway stands a white marble plinth with one digitalrushi gift box on it, lid open, exactly matching the box reference, lit by a soft warm spotlight from above. A few marigold petals drift in the air. Everything else stays identical. No people, no extra text, no watermarks.
Five short clips: fast, then close, then slow, then the arrival, then the hand-over. Each card tells you which start frame to upload, and Videos 2 and 4 also use an end frame.
Start frame: Image 2. About 4 seconds.
The first frame is exactly the start image. The camera is a wide static aerial shot. The six-carriage glass train races along the curving elevated track at extreme speed, heading toward the camera, with heavy motion blur on the carriages and long streaks of warm light trailing along the track. The diya flames flicker in the rush of wind. The town, river, rooftop family and sky lanterns stay still. The train never changes length or shape, and the boxes inside never change. No text.
Start frame: Image 2. End frame: Image 3. About 4 seconds.
The first frame is exactly the start image and the last frame is exactly the end image. The camera dives smoothly toward the moving train while the train slows from high speed to a gentle glide, until the lead carriages fill the right side of the frame and the boxes and sweets inside are clearly visible through the glass roof. The motion blur fades as the train slows. The camera does not rotate. The train keeps its six carriages, and the sweets and boxes stay exactly the same. No text.
Start frame: Image 3. About 4 seconds.
The first frame is exactly the start image. Slow motion. The train glides very slowly while the camera drifts gently along its transparent glass roof, showing the rows of laddoos, kaju katli and maroon digitalrushi gift boxes on the lit shelves inside. Reflections of the sky lanterns slide across the glass and the diya flames flicker softly along the track. The sweets and boxes never move on the shelves. No people inside the train. No text.
Start frame: Image 4. End frame: Image 5. About 5 seconds.
The first frame is exactly the start image and the last frame is exactly the end image. The train glides the last few metres into the decorated station and stops gently. After a beat, the nearest glass door slides open and reveals the digitalrushi gift box on a white marble plinth in the doorway. Marigold petals start drifting down from above. The camera stays completely still. The box design matches the end image exactly. No people. No text.
Start frame: Image 5 (or the last frame of Video 4). About 6 seconds.
The first frame is exactly the start image. The camera stays still. A young college student in a grey hoodie with a backpack on one shoulder walks into the lower left of the frame from behind the camera, seen from behind over his shoulder so his face is never visible. He reaches out with both hands, gently lifts the digitalrushi gift box off the marble plinth and holds it close to his chest. As he takes it, the burnt-orange stripe along the train glows brighter and pulses softly twice, as if the train is greeting him, and the diyas on the platform flare warmly. Marigold petals keep falling. Natural hands with five fingers each. The box keeps its exact design. No text.
Tip: add the words in your editor, not in the AI video — generated text and speech come out garbled. In our reel, “Diwali Sweet Train” runs over Videos 1–3, and Video 5 gets a station-style announcement: “Ghar ki mithai has arrived. Happy Diwali!” (CapCut text-to-speech works). The three things most likely to go wrong are (1) the train changing length or turning up empty — regenerate and keep “six carriages” in the prompt; (2) the box design drifting — attach Image 1 again; (3) the student’s hands in Video 5 — if they still look wrong after 3 tries, end the reel on Video 4 with the box on the plinth.
Start Here
These are the exact tools I use with every client. Start with the Audit — most brands find 3+ leaks on the first pass.
Find exactly what's killing your ROAS in 5 minutes. 5-section framework covering Campaign Structure, Targeting, Creatives, Funnel, and Tracking.
Enter your spend, revenue, and margins. Instantly calculate your true return on ad spend and the break-even ROAS you need to hit profitability.
The exact hook-body-CTA structure I use to write high-converting ad scripts. Plug in your product, follow the formula, watch your CTR climb.
Full Library
Checklists
5-section audit: Campaign Structure, Targeting, Creatives (the big one), Funnel & Tracking. Every unchecked box is money leaking out.
Calculators
Your true ROAS after margin, shipping and RTO — and the break-even number your ads actually need to beat.
Frameworks
Pick a hook, fill four boxes, get a finished script. Built on the hook-problem-proof-CTA structure.
Your ads are leaking money. Every unchecked box is a gap in your funnel — and a direct hit to your ROAS. The good news? It's fixable. DM "AUDIT" on Instagram and I'll personally review your setup for free.
Book a free strategy call and let's scale your brand. No pitch. No fluff. Just an honest look at what's killing your ROAS and how to fix it.