Beyond Scale retweeted
A meteor forty times taller than the tallest tower is falling, and one woman is walking toward it.
The Rotting Hive, Part I.
Created by Seedance 2.5 on @makeugc
On a swamp planet, the Velari grew a living city from flesh, chitin and bone.
They believed nothing from the sky could ever reach them.
Then the meteors came, and the city began to burn.
No one noticed Auriel in her ash-grey robe, or the ancient light she keeps hidden beneath it.
Tonight she finally stops walking.
Part II is already in the works.
The Rotting Hive is only getting started… Part II is where things get interesting. You'll want to be here when it drops.
🤝 Paid partnership · 🤖 Made with AI
This AI video is an absolute masterpiece.
GTA 6 vs love - the answer is obvious🎮🩷
Created by Seedance 2.5 on @makeugc
Workflow + prompt:
- Download a video reference
- Create 3 photo references
- Add the PROMPT:
FORMAT
25-second VERTICAL 9:16 cinematic 3D animated comedy in Pixar style.
Adapt the supplied reference video's narrative, movements, emotions, reactions, camera work and music using exactly THREE principal characters.
Render the entire video in a cohesive Pixar-style animated aesthetic, including characters, escalators, shopping mall, props and living-room fantasy.
REFERENCES
video1 = source choreography, expressions, emotional progression, camera work, editing rhythm and original music, subject to the three-character adaptation below.
image1 = the heavy, balding PS5 shopper on the LEFT escalator.
image2 = the woman in the couple on the adjacent escalator.
image3 = her male companion.
These are the only three character image references.
Preserve each image's exact animated face, hairstyle, physique, clothing, footwear, accessories and colors.
Use the images for identity and wardrobe, not frozen poses.
Transfer the corresponding source performances onto these characters.
THREE-CHARACTER ADAPTATION
The heavy shopper from image1 is the ONLY featured shopper on the left descending escalator.
Remove the source video's second, slim shopper, positioned higher up and behind the heavy shopper on the same left escalator. Remove his PS5 box, shopping bag and separate passage completely.
Do not replace him with another person.
Remove shots and reactions that depend exclusively on his presence.
All of the couple's man's purchase-related attention is directed toward the heavy shopper and HIS PS5 box and shopping bag.
Preserve the remaining source story, including the living-room fantasy, appliance-box handover, crossing attempt, interruption, seated landing and final comic payoff.
CHARACTERS
image1:
The heavy, balding shopper. Preserve his exact reference design and body proportions. He carries one PS5 Pro / GTA VI box and one doll-window shopping bag.
image2:
The woman accompanying image3. Preserve her exact reference appearance and outfit throughout. Transfer the source woman's gestures, attention shifts, reactions and final interaction to her.
image3:
The woman's male companion. Preserve his exact reference appearance and outfit throughout.
He performs the source man's curious looks, temptation, gaming fantasy, distracted reactions, attempted crossing and embarrassed final response.
The man in the fantasy is also image3, wearing the SAME outfit.
No identity swaps, costume changes or additional principal characters.
VISUAL STYLE
High-quality Pixar-style 3D animation with expressive eyes, appealing facial shapes, carefully groomed hair, detailed fabrics and cinematic lighting.
Match the supplied character images consistently in close-ups, profiles and wide shots.
All environments and objects share the same stylized rendering.
Use clear facial acting while preserving the source emotions and comic timing.
Maintain convincing body weight, natural gestures, balance shifts, grounded feet and subtle secondary motion.
The heavy shopper retains his reference physique. Do not slim him down during movement.
No rubbery limbs, floating bodies or excessive squash-and-stretch.
Normal movement speed. No slow motion, fast-forward or speed ramps.
PS5 BOX AND SHOPPING BAG
In the real mall scene, the heavy shopper carries exactly:
- ONE rigid PS5 Pro retail box.
- ONE black “Be happy” shopping bag.
The front of the console box clearly displays:
- “PS5 Pro” branding.
- A white PS5 Pro console product image.
- Grand Theft Auto VI cover artwork.
- Recognizable “grand theft auto” lettering and the Roman numeral “VI”.
The game is GTA VI, meaning GTA 6. Never GTA V.
The artwork is physically printed on the packaging.
Preserve the prop designs from image1, subject to this explicit branding requirement.
The bag has a real oval window cut through its front panel.
Inside is the three-dimensional manufactured doll head from image1.
The doll's face is recessed behind the panel, visible through the opening.
Show cut edges, interior depth, occlusion and shadows.
The doll face is NOT printed on the bag.
It remains an inanimate object with a fixed expression.
No blinking, talking or independent movement.
Follow the source heavy shopper's hand assignments and carrying choreography.
Hands grip the box and bag handles securely.
The box stays rigid. The bag swings subtly with his movement.
Do not create a second PS5 box or second shopping bag in the real mall scene.
APPLIANCE BOX
Preserve the couple's tall green-and-white household-appliance box from video1, rendered in the same animated style.
Keep its source carrying positions and visible handover before the man's crossing attempt.
This is a household-appliance box, not a PS5 box.
During the handover, the woman's receiving grip becomes secure before the man releases it.
She continues supporting the box through the final interaction.
No floating box, unsupported weight or disappearing prop.
ENVIRONMENT
A contemporary shopping mall with adjacent escalators moving in opposite directions.
In the opening composition:
- The heavy shopper occupies the LEFT descending escalator.
- The couple rides the adjacent escalator in the opposite direction.
Preserve the source escalator slope, central divider, rail positions and spacing required for the action.
Maintain coherent screen geography across camera angles.
Do not mirror the layout or reverse the escalators.
Slightly refresh decorative details:
Pale stone finishes, revised ceiling fixtures, different plant pots and subtle changes to distant storefronts.
Do not alter structural elements needed for the blocking.
Keep the architecture and decorations consistent between shots.
Render glass, polished metal, rubber handrails and moving steps in the same Pixar-style aesthetic.
Use warm indoor lighting, soft bounce light, contact shadows and restrained reflections.
Keep the action area clear.
Do not place another featured shopper where the removed source character stood.
The fantasy takes place in a cozy animated living room.
Preserve the source sofa, practical lamp, plants and gaming setup sufficiently to reproduce the original composition and performance.
STORY AND PERFORMANCE
Follow the remaining source beats in this order:
1. ESTABLISH THE THREE CHARACTERS
Show the heavy shopper on the left escalator carrying his PS5 Pro / GTA VI box and doll-window bag.
Show the woman and her companion on the adjacent escalator.
Adapt the source opening composition to these three characters.
2. THE PURCHASES ATTRACT ATTENTION
Reproduce the heavy shopper's satisfied glances and source performance with his purchases.
The couple's man notices the PS5 box and bag.
Match his attention shift, temptation and expression changes.
Preserve the woman's corresponding reactions.
Every purchase-related eyeline points toward the heavy shopper or his props.
No reactions toward an absent second shopper.
3. LIVING-ROOM FANTASY
Retain the imagined gaming sequence.
The image3 man sits on the sofa holding the controller.
Reproduce the source posture, hand movements and pleased expressions.
Keep his exact reference appearance and clothing.
Preserve the relevant fantasy shopping props, using PS5 Pro / GTA VI branding and the same physical doll-window construction.
This is an imagined scene, not a physical transfer of the shopper's possessions.
No additional person appears in the fantasy.
Match the source transition into and out of this sequence.
The fantasy is a brief subjective moment and does not require the same amount of real-world time to pass on the escalators.
4. RETURN TO THE COUPLE
Return to the mall at a spatially coherent point after the initial encounter.
The heavy shopper continues descending naturally.
The woman tries to regain her companion's attention, following the source performance.
He responds with the source distracted reaction.
Omit the second shopper's passage completely.
Continue toward the appliance-box handover without a replacement encounter or prolonged empty staring.
5. HANDOVER AND CROSSING ATTEMPT
Reproduce the appliance-box handover.
Once the woman securely supports it, the man attempts to cross the central divider toward the shopper's escalator.
Match the source preparation, hand placement, weight shift and leg lift.
Maintain visible support and coherent relationships between his body, the divider and moving steps.
His motivation is the heavy shopper's purchases.
Do not make the shopper stop, reverse or reappear at an earlier position.
Do not invent another close pass.
6. INTERRUPTION AND SEATED LANDING
Reproduce the source woman's interruption and the man's resulting return onto the steps.
Use video1 to determine the precise contact and choreography.
Keep the decisive contact, loss of balance and seated landing visible and readable.
Show convincing animated weight and inertia.
Do not substitute a different shove, strike, jump or fall.
Do not replay the contact after a cut.
7. FINAL COMIC PAYOFF
The man remains seated on the moving steps.
The woman stands holding the appliance box.
Preserve her final box gesture and his defeated reaction.
End on the source's final emotional beat.
Do not make him stand again or add a different ending.
The heavy shopper may naturally travel out of frame.
DURATION, CAMERA AND EDITING
Exactly 30 seconds.
Preserve the original order and natural speed of the retained actions.
Adapt the source edit to remove the second shopper while keeping the complete fantasy and final physical sequence.
Use natural-length retained performances and reactions to sustain the duration.
No frozen poses, stretched movement, duplicate actions or artificial pauses.
Match the source camera angles, shot sizes, movement and transitions wherever compatible with the three-character adaptation.
Adjust framing only to maintain clear composition, correct eyelines and visible props.
Keep the handover, crossing preparation, decisive contact and seated landing readable.
Maintain screen direction across cuts.
AUDIO
Use the original music from video1.
Preserve its recognizable rhythm, mood, playback speed and pitch.
Fit it to the 25-second edit with clean, beat-aligned trims.
Retain synchronized sounds and vocal reactions belonging to the three remaining characters.
Remove sounds tied exclusively to the deleted shopper's actions.
Align action sounds with their corresponding contacts.
No added dialogue, narration, lyrics or replacement music.
CONTINUITY
Exactly three principal characters: image1, image2 and image3.
In the real mall scene:
- image1 owns one PS5 Pro / GTA VI box and one doll-window bag.
- The couple carries one separate household-appliance box.
- Ownership changes only through the visible appliance-box handover.
The fantasy does not alter possession of objects in the mall.
Keep identities, proportions, outfits and packaging consistent.
The doll head stays inside its bag and remains inanimate.
Escalator steps, handrails and riders move coherently.
Feet remain supported until a deliberate step or lift.
Bodies and props do not pass through rails, glass or each other.
After the man lands seated, he remains seated through the ending.
NEGATIVE CONSTRAINTS
No second PS5 shopper.
No fourth principal character.
No duplicate heavy shopper or duplicated purchases.
No original live-action faces or costumes.
No mixed animation styles.
No GTA V branding.
No printed doll-face substitute for the physical bag window.
No animated doll head.
No disappearing props, extra limbs, fused fingers or deforming boxes.
No reactions toward absent characters.
No repeated falls, invented collisions or new ending.
No slow motion, speed ramps, freezes or unnatural pauses.
No subtitles, captions, watermarks or end cards.
🤝 Paid partnership · 🤖 Made with AI
🎮 GTA 6 and a PS5 vs love, commitment and married life
Only one of them is coming out on Thursday, and it's not the marriage.
Made with Kling 4.0 on @makeugc
Beyond Scale retweeted
Made with Seedance 2.5 on @makeugc (native 15-second video, no editing)
PROMPT ⬇️
**CAMERA:**
DV 16mm tape camcorder handheld feel. A self-filmed gym vlog: MILA holds the camera for the opening selfie, then props it securely on a low bench for the jump-rope shots. Slight hand shake, misaligned framing, delayed autofocus and a briefly cropped forehead in the selfie. Once propped, the camera stays still; tighter inserts are simple crops of that same recording. Casual hard cuts omit setup time. The camcorder never appears on screen or in the mirror.
**LOOK:**
Soft, slightly blurry tape quality, faint tape noise, bloomed highlights under gym lighting, subtle auto-exposure fluctuations, muted contrast, realistic skin tones. MILA looks like a real woman recorded in an ordinary gym: natural skin texture, small imperfections, fine loose hairs and spontaneous facial expressions remain visible beneath the gentle tape softness. Soft overhead lighting creates believable shadows beneath her chin and feet. No beauty-filter smoothing or glossy commercial grading.
**STYLE:**
Playful, self-deprecating gym-vlog tone — initial confidence, a few genuinely good jumps, one tiny mistake and an extremely well-earned water break. Real effort mixed with humor. Short, casual cuts, normal-speed movement and a light, funny finish.
**Character**
MILA — a beautiful American Instagram-style brunette model and performer, 20. Long dark-brown hair in a high ponytail, glowing skin with a light sweat sheen, glossy lips, expressive eyes, soft glam makeup and delicate sparkly stud earrings. Slim athletic figure with a very large, full bust, very full rounded buttocks, a narrow waist, flat stomach and lean arms and legs.
She wears a cobalt-blue supportive sports bra with thin straps and a deep scoop neckline showing cleavage, paired with matching ultra-short fitted athletic shorts. Bare shoulders, arms and midriff. White ankle socks, sneakers off. She speaks with an American accent.
Her ponytail follows each jump with a slight natural delay. Her shoulders stay relaxed, knees soften on landing, and her fitted sportswear moves naturally with her body.
**Setting**
A gym mat area in the evening, with a flat rubber exercise floor, a mirror wall off to one side and other equipment visible in the background. Soft overhead lighting. One clear, half-full water bottle with a white flip-top drinking spout stands beside the bench, well outside the rope's path. One ordinary dark jump rope with two black handles. Enough clear space above and around her for the full rope arc.
**Storyboard (15s, 6 shots)**
1. *(0–2s, handheld arm's-length selfie)*
MILA looks into the lens with an overly confident little smile, holding both rope handles together in her free hand. One loose strand of hair rests against her cheek. The framing dips slightly as she starts lowering the camera toward the bench.
MILA: “This looks easy.”
2. *(2–5s, propped camera, full-body shot)*
Hard cut: the camera is now resting on the bench, and MILA is already jumping a few steps away. Her whole body, both handles and the rope's overhead arc fit in frame. She makes several small, steady two-foot hops, turning the rope with compact wrist movements, elbows close to her sides. The bottle remains visible beside the bench at the edge of frame. Her ponytail bobs behind her.
No dialogue — rhythmic rope taps, soft landings and light breathing.
3. *(5–7s, tighter insert from the same propped recording)*
A simple crop onto her lower legs, white socks and the floor. The rope skims the rubber just ahead of her toes, passes beneath both feet during each small hop, then continues behind her. Her ankles and knees absorb each landing. A little natural motion blur follows the moving rope.
No dialogue. Keep the same jumping rhythm across the cut.
4. *(7–9s, back to the full-body framing)*
The rope catches lightly against her toes on the next pass. Its rotation stops; she settles onto both feet and takes one small recovery step, still holding a handle in each hand. She looks down at the slack rope, then back toward the lens with an amused, mildly offended expression.
MILA, with a breathy little laugh: “Okay. Rude.”
5. *(9–11.5s, propped camera, medium crop near the bench)*
Jump cut: she is crouched beside the bench. She places both handles and the slack rope on the floor, then picks up the same water bottle with her left hand and flips its drinking spout open with her right. Her shoulders rise and fall with slightly quicker breathing; she smiles to herself.
No dialogue. A small plastic click and quiet gym ambience.
6. *(11.5–15s, closer crop, camera still propped)*
Still crouching near the lens, she raises the open bottle, takes one brief, ordinary sip, lowers it and swallows. The water tilts naturally inside the bottle. She lets out a small satisfied breath, looks straight into the camera and gives a cheeky half-smile.
MILA: “Still counts.”
Finish on her amused expression, with the bottle lowered in her hand.
**Audio and continuity**
Natural American-accented speech with accurate lip sync. Soft gym room tone, rope swishes and floor taps synchronized with each rotation, light breathing, the bottle's lid click and a quiet sip. Leave space between actions and dialogue. No music, voiceover, captions or added comedy effects.
Keep MILA's face, proportions, outfit, hairstyle, lighting and surroundings consistent. The rope remains one continuous cord attached to the same two handles: one rotation per hop, no extra loops, disappearing sections or passage through her body. Both hands hold their respective handles throughout the jumping. The same bottle stays beside the bench until she visibly picks it up.
#AIVideo
🤖 Made with AI
Made with Seedance 2.5 on @makeugc (native 15-second video, no editing)
PROMPT ⬇️
**CAMERA:**
DV 16mm tape camcorder handheld feel. A self-filmed gym vlog: MILA holds the camera for the opening selfie, then props it securely on a low bench for the jump-rope shots. Slight hand shake, misaligned framing, delayed autofocus and a briefly cropped forehead in the selfie. Once propped, the camera stays still; tighter inserts are simple crops of that same recording. Casual hard cuts omit setup time. The camcorder never appears on screen or in the mirror.
**LOOK:**
Soft, slightly blurry tape quality, faint tape noise, bloomed highlights under gym lighting, subtle auto-exposure fluctuations, muted contrast, realistic skin tones. MILA looks like a real woman recorded in an ordinary gym: natural skin texture, small imperfections, fine loose hairs and spontaneous facial expressions remain visible beneath the gentle tape softness. Soft overhead lighting creates believable shadows beneath her chin and feet. No beauty-filter smoothing or glossy commercial grading.
**STYLE:**
Playful, self-deprecating gym-vlog tone — initial confidence, a few genuinely good jumps, one tiny mistake and an extremely well-earned water break. Real effort mixed with humor. Short, casual cuts, normal-speed movement and a light, funny finish.
**Character**
MILA — a beautiful American Instagram-style brunette model and performer, 20. Long dark-brown hair in a high ponytail, glowing skin with a light sweat sheen, glossy lips, expressive eyes, soft glam makeup and delicate sparkly stud earrings. Slim athletic figure with a very large, full bust, very full rounded buttocks, a narrow waist, flat stomach and lean arms and legs.
She wears a cobalt-blue supportive sports bra with thin straps and a deep scoop neckline showing cleavage, paired with matching ultra-short fitted athletic shorts. Bare shoulders, arms and midriff. White ankle socks, sneakers off. She speaks with an American accent.
Her ponytail follows each jump with a slight natural delay. Her shoulders stay relaxed, knees soften on landing, and her fitted sportswear moves naturally with her body.
**Setting**
A gym mat area in the evening, with a flat rubber exercise floor, a mirror wall off to one side and other equipment visible in the background. In shots 2–4, the camera looks diagonally toward the mirror so MILA's full-length reflection is clearly visible beside her, matching her movements and the rope on every turn, including the criss-cross jump. Soft overhead lighting. Across the aisle, one gym member pedals a stationary bike and another does seated cable rows. Beside the bench, well outside the rope's path, stands a half-full, store-bought 500 ml bottle of water. It has thin clear ribbed plastic, a narrow neck and a small white screw cap. One ordinary dark jump rope with two black handles. Enough clear space above and around her for the full rope arc.
**Storyboard (15s, 6 shots)**
1. *(0–2s, handheld arm's-length selfie)*
MILA looks into the lens with an overly confident little smile, holding both rope handles together in her free hand. One loose strand of hair rests against her cheek. The framing dips slightly as she starts lowering the camera toward the bench.
MILA: “This looks easy.”
2. *(2–5s, propped camera, full-body shot)*
Hard cut: the camera is now resting on the bench, and MILA is already jumping a few steps away. Her whole body, both handles and the rope's overhead arc fit in frame. She makes several small, steady two-foot hops, turning the rope with compact wrist movements, elbows close to her sides. The bottle stands at the edge of frame. Several metres behind MILA, both gym members are visible at their machines, each absorbed in their own workout. Her ponytail bobs behind her.
No dialogue — rhythmic rope taps, soft landings and light breathing.
3. *(5–7s, slightly tighter full-body crop from the same propped recording)*
On one turn, as the rope swings down in front of her, MILA crosses her forearms low at her waist, right wrist over left. She clears the crossed loop with one small two-foot hop. She opens her arms on the next turn and clears that one cleanly too. Both wrists and both feet stay in frame. A quick, pleased grin.
No dialogue. Two evenly spaced taps of the rope on the rubber floor.
4. *(7–9s, back to the full-body framing)*
The rope catches lightly against her toes on the next pass. Its rotation stops; she settles onto both feet and takes one small recovery step, still holding a handle in each hand. She looks down at the slack rope, then back toward the lens with an amused, mildly offended expression.
MILA, with a breathy little laugh: “Okay. Rude.”
5. *(9–11.5s, propped camera, medium crop near the bench)*
Jump cut: she is crouched beside the bench. She places both handles and the slack rope on the floor, then picks up the same water bottle with her left hand and twists off its loosely fitted cap with her right. The cap stays between her fingers. Her shoulders rise and fall with slightly quicker breathing; she smiles to herself.
No dialogue. The thin plastic crinkles softly in her grip over the hum of the exercise bike.
6. *(11.5–15s, closer crop, camera still propped)*
Still crouching near the lens, she brings the bottle's narrow open neck to her lips, tips it for one small sip, lowers it and swallows. The water tilts naturally inside the bottle. She lets out a small satisfied breath, looks straight into the camera and gives a cheeky half-smile.
MILA: “Still counts.”
Finish on her amused expression, with the bottle lowered in her hand.
**Audio and continuity**
Natural American-accented speech with accurate lip sync. Soft gym room tone, rope swishes and floor taps synchronized with each rotation, light breathing, the soft crinkle of the water bottle and a quiet sip. Leave space between actions and dialogue. No music, voiceover, captions or added comedy effects.
Keep MILA's face, proportions, outfit, hairstyle, lighting and surroundings consistent. The rope remains one continuous cord attached to the same two handles: one rotation per hop, no extra loops, disappearing sections or passage through her body. Both hands hold their respective handles throughout the jumping. The same bottle stays beside the bench until she visibly picks it up.
#AIVideo
Kling 4.0 or FLUX 3 Max? Which one pulled off the bullet-time shot better? 👀
I gave both models the exact same prompt on @makeugc. The woman in red had 15 seconds to walk to the crosswalk, stop the bullets midair, and deliver one line: “Not today.”
Full prompt below 👇
Create a 15-second vertical 9:16 photorealistic action scene on a busy city street at dusk. The lead is a beautiful European woman in her late twenties with a slim waist and a full bust. Her fitted red dress follows the curve of her hips, and its knee-length hem moves with each step. She wears black 12 cm stiletto heels. Her pace is unhurried. Her composure makes the final line land.
0–3 seconds — Front view. Open on a full-length shot of the woman already walking toward the lens. The camera tracks backward at her exact pace, keeping her face and heels in frame. Pedestrians flow past on either side. Each heel clicks against the pavement above the murmur of traffic. Ahead of her, just beyond the camera, a pedestrian signal turns red.
3–6 seconds — The camera circles behind her. In one smooth move, the camera arcs around her right side, briefly catching her profile before settling into a full-length rear view. She keeps walking straight throughout the move. From behind, we see the silhouette of the red dress and the red pedestrian signal together. She reaches the curb and stops. The heel clicks stop with her.
6–9 seconds — The truck. Hold the rear view. A dark box truck pulls up on the street just beyond the crosswalk. Its side door slides open. Three masked men step out carrying automatic rifles. All three are visible as they open fire in a short burst. On the first flash, the woman raises one hand toward them, palm out.
9–12.5 seconds — Time slows. Cut to a side view with the men on the left and the woman on the right, her palm between them. In extreme slow motion, several copper-colored bullets travel left to right toward her. Show them moving across open air, slowing visibly, then coming to a complete stop a hand’s breadth from her palm. They hang at slightly different distances. Behind them, a loose newspaper turns almost imperceptibly in the air. The gunfire stretches into a low rumble and falls silent.
12.5–15 seconds — “Not today.” Cut to a tight front three-quarter close-up from the street side. Suspended bullets catch the red signal light in the foreground. Shift focus past them to her eyes. Keeping her palm raised, she turns her head the last few degrees and looks directly into the lens. In a calm, dry voice, at normal speed, she says, “Not today.” Hold eye contact for half a beat. Cut to black with the bullets still hanging in place.
🤝 Paid partnership · 🤖 Made with AI
Survived a jarl and a giant. Then she found my feet.
The Saga of Astrid, Part VI.
Made on @makeugc (Seedance 2.5). Details below 👇
Everyone checks faces to spot AI. Look at the pause instead.
In most AI videos nothing ever stops. Everyone moves at the same speed all the time, so a joke has no room to land.
Here the door bangs open and the room just freezes. Two women sit up straight with stone faces, still red from laughing, and for a second nobody moves. He stares at them, turns around and leaves without a word.
Same with the start. Her mouth pops open, she clamps it shut, cheeks puffed, eyes wide, fighting a laugh she clearly can't hold. Holding it in is much harder to fake than letting it out.
Part VII is already in the works.
What would you have done in his place?
🤖 Made with AI
He waited until I took my armor off. Bad plan.
The Saga of Astrid, Part V.
Made on @makeugc (Seedance 2.5). Details below 👇
Everyone checks faces to spot AI. Look at the water instead.
In most AI videos water is decoration. A splash hangs in the air like glass, and a second later everyone is dry again with perfect hair.
Here nothing dries. Her braids stay soaked from the first frame to the last. The bucket breaks over his head and the water actually lands on him and crashes to the floor. When he goes into the tub, half of it ends up on the boards.
Same with the start. Her cheek pressed into wet floorboards, his hand on her head, firelight shaking in the puddle right in front of her face. Wet and messy is much harder to fake than clean.
Part VI is already in the works.
Did Sigrid save her, or just show up for the cleanup?
Beyond Scale retweeted
Jason was NOT prepared for what Lucia had planned for him tonight.
Another hilarious GTA VI experiment made with @makeugc
Lucia decided to surprise Jason with a little private performance. Everything was going perfectly... until it wasn't.
Let's just say this is one night neither of them will ever forget.
Rockstar, we need more moments like this in GTA VI
#AIVideos
🤝 Paid partnership · 🤖 Made with AI
Lucia just made a decision she's going to regret for the rest of her life.
Another GTA VI comedy experiment created with WAN 3.0 on @makeugc
All she wanted was a new tattoo on her back. Simple enough, right?
Well, apparently trusting the wrong tattoo artist in Vice City can have some pretty serious consequences.
Just wait until she sees the result.
#AIVideos
Beyond Scale retweeted
POV: you found the mermaids at your Miami hotel pool trying to dive
Tested Seedance 2.5 on @makeugc. Costs $1 to try it yourself, setup in the replies 👇
Night 13. Turns out silicone floats.
They've been trying to go under since 2am.
Tomorrow you're helping them. How do they get down?
Prompt:
Amateur vertical iPhone video, 9:16, filmed at 2am at a Miami hotel pool with the phone's harsh built-in flash, shaky handheld, grainy, slight motion blur, auto-exposure flicker, candid real home video look, not cinematic, no music. The man filming is at the pool edge; the bright lit turquoise pool water is clear enough to see the mermaids' faces, bodies and fish tails under the surface; a shiny metal pool ladder goes down into the water at the left edge of the frame, right next to the LEFT one, and it stays visible at the left edge of the frame in the close shots too; behind them the far side of the pool, palm trees and dim hotel lights. The three mermaids stay side by side in the same places the whole video.
The creatures: three real mermaids with shimmering fish tails clearly visible under the lit water from the very first frame. All three are glamorous Miami women with hugely overfilled, giant, glossy duck lips, so swollen and stiff they can barely open their mouths, long fake lashes, heavy contour makeup, long stiletto acrylic nails. Their giant lips are so full of silicone that they float: whenever a mermaid goes under, her lips pull her face back up to the surface like a cork, lips first. The lips are always attached to their faces, the same real lips, never separate objects floating on the water. Three clearly different women, not twins:
LEFT: tanned Latina with long black wet hair, giant nude glossy duck lips, gold-and-bronze tail.
CENTER: platinum blonde with sleek wet hair, giant pink glossy duck lips, baby-blue tail.
RIGHT: pale redhead with wet copper curls, giant red glossy duck lips, emerald-green tail, sly and quiet.
Hands: the man never puts his hand in the frame. Each mermaid has only her own two hands; no extra hands appear.
Important: the mermaids don't say any words, only muffled grunts of effort through their stiff lips. Nobody laughs at any point. Their eyes are open and alive, even under the water. All three are alive and moving every second: no one ever freezes while another acts. Their movements and reactions are always different and asynchronous, never mirrored: each girl does her own thing.
0-1.5s: From the very first frame the man is standing at the pool edge filming down at the water from about two meters: on the lit turquoise surface only three pairs of giant glossy lips stick out of the water, pink, nude and red, bobbing like three pool floats; the faces are right under the surface, wet hair spreading around them, and the three shimmering tails are visible deeper in the water. The man whispers, stunned: "What..." and quickly kneels down, the camera moving in close, about half a meter from them, while the three of them lift their heads up out of the water one after another, not at the same time, and look at him.
1.5-4.5s: Close: three faces big in the frame, wet, giant lips glistening. The CENTER one dives head-first hard, her baby-blue tail flipping high up into the air; her body goes down under the water, but after a second her giant pink lips shoot back up to the surface like a cork and drag her face up with them, splash. She surfaces up to her nose, eyes furious. The other two watch, tails flicking.
4.5-8.5s: The LEFT one grabs the metal pool ladder next to her with both hands and pulls herself down rung by rung, deeper under the water, her gold tail kicking, her face straining while her lips keep pulling her face up toward the surface; she lets go of the ladder and shoots straight up out of the water, lips first, with a big splash. The CENTER one glares at her, the RIGHT one watches, thinking.
8.5-12s: The RIGHT one climbs on top of the other two, pressing down on both their heads with her hands and her emerald tail to push them under; for a moment all three sink together, then all three pop back up to the surface at once, three pairs of lips first, with a big splash toward the camera.
12-15s: They give up and float on their backs at the surface, three pairs of giant lips sticking up out of the water like buoys; the CENTER one rolls her eyes at the camera, the LEFT one slaps the water with her hand, the RIGHT one sulks, and their tails slap the surface in frustration, until the last second.
Sound: pool filter hum, big dive splashes, the hollow pop of lips breaking the surface again and again, the metal ladder creaking, water churning, tail slaps, muffled grunts of effort, the man's stunned whisper and breathing, waves hitting the edge. No laughter. Faces stay lit by the flash the whole time.
🤝 Paid partnership · 🤖 Made with AI
POV: you brought a sushi boat to the mermaids at your Miami hotel pool
Tested Seedance 2.5 on @makeugc. Costs $1 to try it yourself, setup in the replies 👇
Night 12. Back to where it all started.
The redhead still doesn't share.
Tomorrow you're feeding them. What's on the menu?
Prompt:
Amateur vertical iPhone video, 9:16, filmed at 2am at a Miami hotel pool with the phone's harsh built-in flash, shaky handheld, grainy, slight motion blur, auto-exposure flicker, candid real home video look, not cinematic, no music. The man filming is kneeling on the pool edge, camera low and close, about half a meter from their faces. Layout locked for the whole video: the three mermaids are in the water right in front of him, leaning on the same pool edge; a big wooden sushi boat sits low on the edge between the RIGHT mermaid and the camera, at the right side of the frame, loaded with salmon nigiri, tuna rolls, shrimp tempura rolls and a pile of salmon sashimi; behind them the bright lit turquoise pool water, clear enough to see into, the far side of the pool, palm trees and dim hotel lights.
The creatures: three real mermaids leaning on the pool edge with their forearms, close to the lens, their shimmering fish tails visible under the lit water behind them. All three are glamorous Miami women with hugely overfilled, giant, glossy duck lips, so swollen and stiff they can barely open their mouths, long fake lashes, heavy contour makeup, long stiletto acrylic nails. Three clearly different women, not twins:
LEFT: platinum blonde with sleek wet hair, pink glossy lips, baby-blue tail.
CENTER: tanned Latina with long black wet hair, nude glossy lips, gold-and-bronze tail.
RIGHT: pale redhead with wet copper curls, red glossy lips, emerald-green tail, sly and quiet.
Important: they speak slowly, slurred and muffled, barely moving their giant lips, as if their mouths can't open. Nobody laughs at any point. All three are alive and moving every second: no one ever freezes or stares blankly while another acts. They glance at each other, at the food and at the camera, shift their weight on the edge, flick their tails under the water, react to each other with real, different, asynchronous movements, like real greedy animals around food.
0-2s: The flash hits all three faces at once, faces big in the frame. The CENTER one leans in closest and says slowly through barely moving lips, muffled: "Feed us." At the same time the LEFT one pushes her shoulder against the CENTER one to get closer to the lens, and the RIGHT one is already eyeing the sushi boat next to her, her tail swishing under the water.
2-5s: The man's hand holds out one big fat salmon roll to the LEFT one. She snatches it and tries to push it into her mouth, but her swollen lips barely part, the roll squashes against them and rice sticks all over her pink gloss and chin. She mumbles, muffled: "It doesn't fit." Meanwhile the CENTER one grabs at the man's now empty hand with her nails, glaring at the LEFT one, and the man pulls his hand back out of the frame. The RIGHT one quietly slides closer to the sushi boat, glancing between the others and the camera.
5-8s: The CENTER one grabs a big handful of salmon nigiri straight from the boat and forces it through her stiff swollen lips, rice spilling down her chin, her cheeks ballooning as she chews hard. The LEFT one, trying to pick the rice off her lips but only smearing it more, reaches for the boat too; the CENTER one shoves her hand away and splashes water at her with her tail. The RIGHT one watches them fight and slowly hooks the edge of the sushi boat with one long nail.
8-11s: The RIGHT one slides the whole sushi boat off the pool edge into the water and sinks under with it, holding it to her chest, a few rolls floating on the surface, her emerald tail flipping up and slapping the water.
11-15s: The LEFT and CENTER ones notice too late: both turn their heads sharply, look at each other with rice on their lips, then both dive under after her at the same moment, their tails flipping up out of the water and slapping the surface, splashing water at the camera, the floating rolls bobbing in the waves, until the last second.
Sound: pool filter hum, water splashes, wet chewing, muffled slurred voices, rolls squishing, the wooden boat scraping on the tiles, a splash as the boat goes under, loud tail slaps, waves hitting the edge, the man's quiet stunned breathing. No laughter. Faces stay sharp, close and lit by the flash the whole time.
Beyond Scale retweeted
Wanna cut the AI slop out of your videos as easily as she cuts through zombies? 🧟
The 12-minute UGC guide in the quoted post shows you how.
Made the zombie clip on @makeugc
🤖 Made with AI
Your AI UGC ads look and sound fake. But they don’t have to.
We just dropped a full video showing how to create AI UGC videos that actually look realistic to use for your ad campaigns.
RT + comment and we’ll send you 300 free credits 👇
youtu.be/5Kd56IFo4o8?si=usiO…
Рe cut through a steel column like it was paper. Then he got serious
Created by Seedance 2.5 on @makeugc
The molten metal isn't the twist. Watch the last 2 seconds, then watch them again.
Prompt:👇
FORMAT: 23 seconds, 16:9, live-action dark action thriller.
CINEMATIC QUALITY (applies to every shot):
Must look like real footage from a top-budget live-action feature film, indistinguishable from a real camera shoot, never like AI. Shot on ARRI Alexa 65 with sharp high-end prime lenses, cinematography in the style of Greig Fraser: motivated natural light, soft realistic shadows, true-to-life exposure, deep but detailed blacks, no blown highlights. 24 fps with natural 180-degree shutter motion blur. Native high resolution with razor-sharp focus on the subject, crisp micro-contrast and high-frequency detail on every edge and surface. Subtle organic 35mm film grain over the whole image. Natural, slightly muted real-world color grade with accurate skin tones.
TEXTURE REALISM:
Every surface has real physical texture: visible skin pores, fine facial hair, natural skin imperfections and freckles, individual hair strands, fabric weave and seams, scratches, dust, dirt, wear and moisture on materials, real reflections on wet or polished surfaces. Water lies on surfaces as droplets, puddles and wet sheen rather than as haze in the air. Atmosphere stays clear enough that background detail remains sharp.
MOTION REALISM:
All movement looks performed by real actors and stunt performers: real weight, inertia, momentum, balance and recoil, feet planted on the ground, natural acceleration and deceleration. Every physical contact is visible and has an immediate physical reaction. Camera movement is operated by a real crew: steady, motivated by the action, with a short natural jolt only on heavy impacts.
ANTI-AI RULES:
No motion smoothing, no frame interpolation, no beauty filter, no skin retouching, no waxy or plastic skin, no airbrushed surfaces, no over-sharpened halos, no smeared or melting details in fast motion, no floaty weightless movement, no morphing faces or objects, no oversaturated colors, no fake glow or bloom, no video game or CGI render look, no low resolution, no compression blur.
LIGHT EFFECTS EXCEPTION: The no-glow, no-bloom, no-oversaturation and muted-grade rules apply to everything except the story's own light effects. Power beams, the glowing suit, energy and explosions may glow intensely in their own colors and must behave like real light sources: they cast colored light, reflections and moving shadows on faces, debris and surroundings, and never look like flat overlays.
ACTION ATMOSPHERE: Smoke, dust, ash, sparks and debris produced by the action are allowed and physically driven, staying local to their sources; the main action and background events stay clearly readable.
REFERENCES:
Image 1: Akane, defines face, hair and outfit. Image 2: the Knight, defines armor, helmet and sword.
CHARACTERS:
AKANE: Japanese woman, 24, athletic, 168 cm. Black hair in a high ponytail, thin scar through her left eyebrow. Matte dark-navy tactical bodysuit with scuffed armored shoulder pads, black half-mask over the eyes, two short straight blades in back sheaths, a steel grappling cable on her belt. Fast, aggressive, precise. Same face, hair and outfit in every shot; damage accumulates and stays.
THE KNIGHT: 1.3x Akane's height (about 220 cm). Forged dark-iron helmet with a narrow eye slit, notched plate armor, torn rust-brown cloak, two-handed sword 1.6 m long. Slow, calm, every blow heavy enough to cut steel. Weak point: a gap in the armor under his left arm. One design and one scale for the whole film.
ENVIRONMENT + SCENE LAYOUT:
Abandoned steel foundry, daytime. Rusty steel beams, broken skylights, a single thick shaft of dusty daylight falling on the center of the concrete floor. Fixed layout from frame 1: an upper steel beam 8 m above the floor running left to right; THE KNIGHT standing motionless in the light shaft at center-right; a rusty support column LEFT of center; heavy hanging chains and a crane hook on the LEFT; a stack of steel sheets on the LEFT behind Akane's fighting position; a large foundry ladle hanging on a single steel cable on the RIGHT behind the Knight, its lip glowing with leftover molten metal. Dust floating in the light, rust flakes, scattered scrap on the floor. No other characters.
POWERS & ENEMY:
No superpowers, pure steel, weight and skill. Akane's speed versus the Knight's mass. Escalation: blade strikes, the environment, then the molten metal. The Knight survives everything and keeps coming.
TIMELINE:
0-3s HEROINE: swings down from the upper beam on her steel cable, frame-left to frame-right, draws her right blade mid-swing, eyes locked on the Knight, strikes the right side of his helmet and rolls to a landing on frame-left. ENEMY: the blade glances off the helmet with sparks, his head jerks. BACKGROUND ACTION: dust falls from the beam. CAMERA: 3/4 front tracking her swing, the strike fully visible. END STATE: Akane crouched 4 m frame-left of him, facing right.
3-6s ENEMY: turns and swings his sword horizontally at her neck height. HEROINE: drops flat on her back, the blade passes inches above her face. FX & DEBRIS: the sword slices clean through the rusty column behind her; its top half crashes down in dust and sparks. CAMERA: low side angle. END STATE: Akane back in a low stance with both blades.
6-9s HEROINE: four fast strikes at helmet, shoulder, knee, forearm. ENEMY: blocks three; the fourth cuts a deep notch in his right vambrace; he answers with a pommel strike to her face, her mask cracks, she staggers back toward the steel sheets. CAMERA: medium 3/4 front, continuous. END STATE: Akane frame-left by the sheets.
9-12s ENEMY: overhead chop. HEROINE: sidesteps right; the sword cuts the top steel sheets in half with a shower of sparks. She ducks under his left arm and drives her left blade into the armor gap under it. ENEMY: roars and smashes his elbow into her chest. CAMERA: medium side angle, the thrust into the gap clearly visible. END STATE: Akane thrown backward toward the chains.
12-15s HEROINE: slams back into the hanging chains, rebounds, lands on her feet, breathing hard, one blade left in her right hand. ENEMY: pulls her blade out of his side and throws it clattering to the floor; smoke drifts from the gap. BACKGROUND ACTION: chains swinging, rust raining down. CAMERA: wide side angle, jolt on impact. END STATE: Akane frame-left, Knight center, ladle glowing behind him on the right.
15-18s HEROINE: eyes on the ladle cable, throws her last blade past the Knight. FX & DEBRIS: the blade cuts the cable, the ladle tips, a stream of glowing molten metal pours onto the Knight; fire and fountains of sparks, his cloak burns away, his armor glows orange. CAMERA: wide 3/4 shot, short slow motion. END STATE: the Knight engulfed in fire.
18-21s ENEMY: steps out of the fire, armor melted and glowing, sword red-hot. HEROINE: unarmed, charges him. ENEMY: catches her by the throat with one hand, lifts her and hurls her into the stack of steel sheets, which collapse onto her. HEROINE: pushes up on one elbow, mask broken, looks up at him without fear. CAMERA: low angle on him, then follows her flight. END STATE: Akane half-buried frame-left, Knight standing over her raising the glowing sword.
21-23s CAMERA: hard cut to a high top-down wide shot: two figures in the light shaft, the glowing blade starts to fall, and at that instant the daylight shaft goes dark. Black screen for the last 0.5 seconds, only the hiss of cooling metal. The blow itself is never shown.
CAMERA:
Real film crew operation, motivated movement, wide and medium shots that keep both fighters and every contact readable, low angles to show the Knight's scale. Maximum 8 camera setups, one per beat, cuts only on motion. Warm dusty daylight from the skylights against cold rusty steel; molten metal and the red-hot sword are the only glowing sources.
AUDIO & DIALOGUE:
Footsteps on the beam, cable whir, steel on steel, sparks hissing, the column crashing, armor clanking, Akane's sharp breaths and grunts, the Knight's single low roar when stabbed, chains rattling, cable snapping, molten metal roaring and sizzling, steel sheets collapsing. Akane, on the floor, low: "Do it." Final: silence, then the hiss of cooling metal. No music.
CONTINUITY & STAGING RULES:
- SCENE LAYOUT is fixed from frame 1. Every vehicle, creature and prop that appears in this clip is visible or clearly placed from the first second. Nothing new enters the scene unless the timeline shows it entering from a visible point (street end, alley, rooftop).
- No object changes type, position or state unless the timeline causes it. A parked vehicle stays parked in the same spot until it is hit. Damage stays.
- Maximum 8 camera setups in the clip, one per beat. Prefer one continuous tracking shot. No random angle cuts.
- Camera stays on one side of the action (180-degree rule). Heroine stays frame-left, enemy stays frame-right unless the timeline moves them.
- Heroine's face or 3/4 profile and the target are both visible during every strike. No attacks filmed from directly behind her.
- Every strike shows the weapon or power physically touching a named point on the target, with damage appearing exactly at that point. No air swings, no target breaking on its own.
- Monster keeps one design, one head and the same scale (Knight 1.3x heroine height) for the whole clip.
- Heroine stays on the ground unless a jump is described with take-off and landing.
- No text, letters or logos on vehicles, signs or screens.
- One action at a time. Each beat finishes before the next begins.
NEGATIVE CONSTRAINTS:
No looking at camera during attacks, no attacks on unseen targets, no attacks filmed from directly behind, no air swings, no objects appearing from nowhere, no static or frozen background characters, no idle crowds, no direction flips between shots, no background objects resetting after damage, no monster design or scale changes, no text or lettering on vehicles and signs. No killing blow on screen, no gore, no blood pools, no dismemberment, no superpowers, no magic, no extra characters, no duplicated weapons, Akane never gets her blades back once lost, no sexualized framing, no music, no subtitles.
🤝 Paid partnership · 🤖 Made with AI
The most expensive nightclub in shanghai lasted 30 seconds
Created by Seedance 2.5 on @makeugc
Most people never look behind the boss. Second watch, look at the back wall from 0:01.
Prompt:👇
FORMAT: 30 seconds, vertical 9:16 full frame, photorealistic martial arts action.
CINEMATIC QUALITY (applies to every shot):
Must look like real footage from a top-budget live-action feature film, indistinguishable from a real camera shoot, never like AI. Shot on ARRI Alexa 65 with sharp high-end prime lenses, cinematography in the style of Greig Fraser: motivated natural light, soft realistic shadows, true-to-life exposure, deep but detailed blacks, no blown highlights. 24 fps with natural 180-degree shutter motion blur. Native high resolution with razor-sharp focus on the subject, crisp micro-contrast and high-frequency detail on every edge and surface. Subtle organic 35mm film grain over the whole image. Natural, slightly muted real-world color grade with accurate skin tones.
TEXTURE REALISM:
Every surface has real physical texture: visible skin pores, fine facial hair, natural skin imperfections and freckles, individual hair strands, fabric weave and seams, scratches, dust, dirt, wear and moisture on materials, real reflections on wet or polished surfaces. Water lies on surfaces as droplets, puddles and wet sheen rather than as haze in the air. Atmosphere stays clear enough that background detail remains sharp.
MOTION REALISM:
All movement looks performed by real actors and stunt performers: real weight, inertia, momentum, balance and recoil, feet planted on the ground, natural acceleration and deceleration. Every physical contact is visible and has an immediate physical reaction. Camera movement is operated by a real crew: steady, motivated by the action, with a short natural jolt only on heavy impacts.
ANTI-AI RULES:
No motion smoothing, no frame interpolation, no beauty filter, no skin retouching, no waxy or plastic skin, no airbrushed surfaces, no over-sharpened halos, no smeared or melting details in fast motion, no floaty weightless movement, no morphing faces or objects, no oversaturated colors, no fake glow or bloom, no video game or CGI render look, no low resolution, no compression blur.
LIGHT EFFECTS EXCEPTION: The no-glow, no-bloom, no-oversaturation and muted-grade rules apply to everything except the story's own light effects. Power beams, the glowing suit, energy and explosions may glow intensely in their own colors and must behave like real light sources: they cast colored light, reflections and moving shadows on faces, debris and surroundings, and never look like flat overlays.
ACTION ATMOSPHERE: Smoke, dust, ash, sparks and debris produced by the action are allowed and physically driven, staying local to their sources; the main action and background events stay clearly readable.
REFERENCES:
If a heroine photo is provided, it defines her exact face, hair and outfit in every shot. Otherwise follow the description below.
CHARACTERS:
HEROINE: young Chinese woman about 22, petite, flexible, dancer's posture, stunningly beautiful. Straight black hair to the waist, sharp blunt bangs, two thin braids at the temples joined at the back with a gold hair clip. Red lipstick, sharp eyeliner, gold ear cuffs. Outfit: white silk cropped jacket with high mandarin collar and a crimson phoenix embroidered across it, sleeveless with red piping, black leather corset belt over it, crimson pleated mini skirt over black shorts, white knee-high boots with crimson laces, a long crimson silk sash at the waist that streams behind every kick. Calm, cold, confident.
BOSS: bald giant about 1.3x her height, bull neck, black suit, open collar.
GUARDS: about 25 men in black suits and dark sunglasses, some with batons, constantly moving.
GUESTS: about 15 party guests fleeing in the deep background.
ENVIRONMENT + SCENE LAYOUT (fixed from frame 1):
Luxury penthouse nightclub at night, Shanghai skyline glowing through tall windows, red and blue neon, glossy black floor. Camera starts near the entrance looking into the room. Heroine in the center of the dance floor. FRAME-LEFT: tall champagne glass pyramid on a table, behind it a long bar with backlit bottle shelves. CENTER: large crystal chandelier above the dance floor, black leather sofas and glass coffee tables. FRAME-RIGHT: DJ booth with speakers, staircase to a VIP balcony with the boss standing at its top. BACK WALL: huge floor-to-ceiling aquarium with fish and stingrays. Guards in a ring around her, guests near the back exit.
POWERS & ENEMY:
Elite kicker. Before a power kick crimson sparks run along her calves with a low hum for half a second. Kicks leave a crimson-gold arc trail in the air and a visible air shudder on impact. Recoil: she slides half a meter on the floor. Cost: after the final kick she breathes hard. Escalation: plain kicks, spinning kick with energy rings, final phoenix kick. Guards are human-sized. Boss is slow but crushing if he grabs. Damage stays to the end.
TIMELINE:
0-3s HEROINE: center, lowering her right leg after a kick, sash still in the air, eyes on the guard ring. ENEMY: a guard flies across frame to the left and crashes into the champagne pyramid, it collapses in a cascade. BACKGROUND: music cuts, guests scream and run to the back exit. FX: glass and champagne spray. CAMERA: wide, her full body, whole room visible. END: guards closing the ring.
3-6s ENEMY: six guards rush her from all sides. HEROINE: drops to a handstand and spins her legs like helicopter blades, boots hit chests and jaws, crimson energy rings ripple outward. ENEMY: all six fly outward, two smash a glass coffee table. CAMERA: wide slightly high angle, full ring visible, jolt. END: six down.
6-9s HEROINE: back on her feet, faces frame-right, side kick sends a leather sofa sliding into four guards at the DJ booth. FX: booth explodes in sparks, speakers topple, part of the neon goes dark. CAMERA: wide 3/4 front, booth in frame. END: four down at the booth.
9-10s HEROINE: fighting stance center, 3/4 to camera, eyes frame-right on remaining guards. BACKGROUND: guards regrouping, sparks still falling. CAMERA: wide, smooth slide left. END: calm stance.
10-13s ENEMY: eight guards with batons from both sides. HEROINE: flips over one, lands on the bar counter frame-left, runs along it kicking attackers off. FX: bottle shelves behind her burst into glass. CAMERA: wide side tracking along the bar, her face 3/4 visible. END: she drops off the bar to the floor.
13-16s ENEMY: boss comes down the staircase frame-right, catches her ankle mid-kick, swings and slams her back onto a glass table. HEROINE: table shatters, she lies on the floor, pain, no blood. BACKGROUND: guards laugh. CAMERA: wide, boss and her full bodies, heavy jolt. END: she lies center, boss over her.
16-19s HEROINE: on her back, double kick with both boots into boss's chest. ENEMY: he slides 3 m back into the aquarium wall, glass cracks into a spiderweb, water seeps. CAMERA: wide low angle, aquarium filling the background. END: boss stunned at the cracked aquarium.
19-20s HEROINE: rises into stance center, faces the boss, wipes her lip with the back of her hand. BACKGROUND: fish darting in the cracked tank. CAMERA: wide, slow slide right. END: calm stance.
20-23s ENEMY: boss and last ten guards charge together. HEROINE: crimson sparks run up her legs, hum builds, sash lifts upward. CAMERA: wide, whole charge visible. END: charge 3 m away.
23-26s Slow motion on impact. HEROINE: leaps and spins in the air, heel wrapped in crimson-gold fire strikes the boss's chest, fire trail flares in the shape of a phoenix wing. ENEMY: boss blasts back through the aquarium, the wall bursts, a wave of water with fish and stingrays floods the club and sweeps the guards away. CAMERA: wide front 3/4, her face and the boss visible, never from behind, heavy jolt. END: she lands.
26-29s HEROINE: lands on the bar counter, the wave passes beneath her. FX: chandelier drops into the water with sparks, sofas float, guards washed against walls. CAMERA: wide, whole flooded room. END: water settling.
29-30s HEROINE: sits on the edge of the bar, crosses her legs, adjusts her sash, looks into the lens with a slight smile, no attack. BACKGROUND: boss lying on the wet floor below, a fish flapping beside him. CAMERA: slow push from wide to medium. END: final pose.
CAMERA:
FRAMING IS WIDE: 24-28mm lens, camera 6-10 m from her. In every shot before 29s her whole body from head to boots is visible and fills one third to one half of frame height; the rest of the frame shows the room, guards, destruction and background action. No close-ups before 29s. Cuts only on motion, maximum 8 setups. Slow motion only at 23-26s. Vertical 9:16 composition with the room depth behind her.
AUDIO & DIALOGUE:
Club bass cut off abruptly, guests screaming, boot impacts, bodies on glass and leather, glass cascades, spark crackle, low energy hum, deep impact boom, glass wall bursting, roaring water. Boss at 13s: "Got you, little bird." Heroine at 29s, calm: "Phoenix."
CONTINUITY & STAGING RULES:
- SCENE LAYOUT is fixed from frame 1. Every vehicle, creature and prop that appears in this clip is visible or clearly placed from the first second. Nothing new enters the scene unless the timeline shows it entering from a visible point (street end, alley, rooftop).
- No object changes type, position or state unless the timeline causes it. A parked vehicle stays parked in the same spot until it is hit. Damage stays.
- Maximum 8 camera setups in the clip. Prefer one continuous tracking shot. No random angle cuts.
- Camera stays on one side of the action (180-degree rule). Bar and champagne pyramid stay frame-left, DJ booth and staircase stay frame-right unless the timeline moves them.
- Heroine's face or 3/4 profile and the target are both visible during every strike. No attacks filmed from directly behind her.
- Every strike shows the weapon or power physically touching a named point on the target, with damage appearing exactly at that point. No air swings, no target breaking on its own.
- Guards keep human scale, the boss stays 1.3x heroine height for the whole clip.
- Heroine stays on the ground unless a jump is described with take-off and landing.
- No text, letters or logos on vehicles, signs or screens.
- One action at a time. Each beat finishes before the next begins.
NEGATIVE CONSTRAINTS:
No text, captions, titles, subtitles or watermarks anywhere. No letterbox, no black bars, full 9:16 frame. No close-ups, no tight framing, no empty background. No looking at camera during attacks, no attacks on unseen targets, no attacks filmed from directly behind, no air swings, no objects appearing from nowhere, no static or frozen background characters, no idle crowds, no direction flips between shots, no background objects resetting after damage, no boss design or scale changes, no text or lettering on signs and screens. No blood, no wounds. No outfit, hair or face changes.
Beyond Scale retweeted
They brought 200 soldiers and a giant into a valley full of swords. Nobody told them whose swords
Created by Seedance 2.5 on @makeugc
Watch her bracers at 0:07. Her own power almost loses this fight.
Prompt:👇
FORMAT: 15 seconds, 16:9 horizontal, epic fantasy action.
CINEMATIC QUALITY (applies to every shot):
Must look like real footage from a top-budget live-action feature film, indistinguishable from a real camera shoot, never like AI. Shot on ARRI Alexa 65 with sharp high-end prime lenses, cinematography in the style of Greig Fraser: motivated natural light, soft realistic shadows, true-to-life exposure, deep but detailed blacks, no blown highlights. 24 fps with natural 180-degree shutter motion blur. Native high resolution with razor-sharp focus on the subject, crisp micro-contrast and high-frequency detail on every edge and surface. Subtle organic 35mm film grain over the whole image. Natural, slightly muted real-world color grade with accurate skin tones.
TEXTURE REALISM:
Every surface has real physical texture: visible skin pores, fine facial hair, natural skin imperfections and freckles, individual hair strands, fabric weave and seams, scratches, dust, dirt, wear and moisture on materials, real reflections on wet or polished surfaces. Water lies on surfaces as droplets, puddles and wet sheen rather than as haze in the air. Atmosphere stays clear enough that background detail remains sharp.
MOTION REALISM:
All movement looks performed by real actors and stunt performers: real weight, inertia, momentum, balance and recoil, feet planted on the ground, natural acceleration and deceleration. Every physical contact is visible and has an immediate physical reaction. Camera movement is operated by a real crew: steady, motivated by the action, with a short natural jolt only on heavy impacts.
ANTI-AI RULES:
No motion smoothing, no frame interpolation, no beauty filter, no skin retouching, no waxy or plastic skin, no airbrushed surfaces, no over-sharpened halos, no smeared or melting details in fast motion, no floaty weightless movement, no morphing faces or objects, no oversaturated colors, no fake glow or bloom, no video game or CGI render look, no low resolution, no compression blur.
LIGHT EFFECTS EXCEPTION: The no-glow, no-bloom, no-oversaturation and muted-grade rules apply to everything except the story's own light effects. Power beams, the glowing suit, energy and explosions may glow intensely in their own colors and must behave like real light sources: they cast colored light, reflections and moving shadows on faces, debris and surroundings, and never look like flat overlays.
ACTION ATMOSPHERE: Smoke, dust, ash, sparks and debris produced by the action are allowed and physically driven, staying local to their sources; the main action and background events stay clearly readable.
CHARACTERS:
HEROINE: woman around 25, pale freckled skin, long wet copper-red braid, green eyes, sharp determined face. Dark brown quilted leather long coat soaked by rain, worn leather bracers studded with iron plates, dark trousers, mud-covered boots. No weapon in her hands. Calm, focused, heavy breathing after each use of power.
SOLDIERS: about 200 infantry in dented dark iron helmets, iron breastplates and round iron shields, charging on foot in a dense block.
COLOSSUS: humanoid war giant 6x the heroine's height, riveted rust-streaked black iron plates, narrow eye slit glowing dull orange, a horizontal glowing orange seam across its chest plate (its weak point, visible from the first second). Right hand holds a long iron chain ending in a spiked iron hammer head the size of a cart.
ENVIRONMENT + SCENE LAYOUT (fixed from frame 1):
Wide muddy valley at grey dawn, light steady rain, heavy low storm clouds, cold soft daylight. The valley floor is covered with thousands of old rusted swords stuck upright in the mud, remains of an ancient battle.
- HEROINE: frame-LEFT foreground, standing in the sword field 5 m in front of a closed wooden palisade gate of her village.
- COLOSSUS: frame-RIGHT background, 60 m away, right arm extended after throwing its hammer.
- CHAIN HAMMER: in flight from frame 1, flying toward the heroine, chain trailing back to the colossus's hand.
- SOLDIERS: frame-RIGHT, behind and around the colossus, 70 m away, charging right to left with torn grey banners.
POWERS & ENEMY:
POWER (magnetism): blue-white electric arcs crawl over her palms, forearms and iron-studded bracers for 0.5 s before each use, with a deep metallic hum. Blades in reach tremble, shed rust dust, then fly along her hand's direction. Recoil: her boots slide back in the mud, braid and coat whip backward. Flaw: her field also grips nearby iron, including the enemy chain. Cost: one thin drop of blood from her nose after the big lift. Level 1: one row of swords. Level 2: every sword in the valley fused into one spear.
COLOSSUS: slow heavy steps that splash mud and shake the swords, attacks with its chain hammer: throw and hard yank back. Armor dents stay.
TIMELINE:
0-3 s (HOOK): The hammer, already in flight at frame 1, rushes at her. She watches it, pushes off her right foot and rolls toward frame-left; the hammer slams into the mud where she stood, ploughing a trench and throwing a fan of rusted swords, mud and water into the air. Soldiers charge behind the colossus. Camera opens low behind the flying hammer, cuts on motion to a low side angle as she rolls clear, strong jolt at impact. END: she kneels frame-left, hammer half-buried 3 m to her right.
3-6 s: She rises, eyes on the front rank, sweeps her right arm toward frame-right. A row of about 50 swords rips out of the mud and slams flat into the front rank's round iron shields with sparks; the rank is knocked back into the mud, the charge stalls. The colossus wraps the chain around its fist. Medium 3/4 front angle, short jolt on impact. END: arm extended, boots slid back 1 m, arcs still on her bracers.
6-9 s (CRISIS): Her charged bracers snap onto the taut chain. The colossus yanks it with a full-body pull and drags her on her knees toward frame-right, ploughing a furrow through the swords, sparks where the bracers grind the chain. She rips her arms apart, the arcs break, she stops. Low tracking shot moving with her, the colossus growing ahead. END: she kneels 20 m in front of the colossus, mud on her face, eyes up at it.
9-12 s (WOW SHOT): She stands, feet wide, raises both arms straight up, eyes on the chest seam, arcs to her shoulders, a drop of blood from her nose. Every sword in the valley tears out of the mud and spirals up into one giant turning ring of steel under the storm clouds. The colossus freezes with its hammer raised, soldiers stop and look up, banners drop. Low wide angle from behind her shoulder at 3/4, her small figure lower left, the ring filling the sky, slow tilt up, the only slow motion in the clip.
12-15 s (CLIMAX): She clenches both fists above her head, the ring fuses with a deafening screech into one huge rusted spear 10x her height, then she throws both arms toward frame-right. The spear drops from the clouds and drives exactly into the glowing chest seam; the plate splits in a burst of orange sparks and the colossus falls backward onto the mud. Soldiers drop shields and run. Medium wide from her side, strong jolt as it falls. Last second: sparks fade, she lowers her arms and stands still frame-left, looking at the fallen giant, rain on her face.
CAMERA:
Maximum 4 setups, cuts only on motion. Low angles keep the colossus scale against her.
AUDIO & DIALOGUE:
Rain, whistle of the incoming hammer, metallic hum building with each use of power, ringing of thousands of blades, shield impacts, chain grinding, iron groan of the giant, final screech of fusing steel. One line at 12 s, low through her teeth: "Not one step further."
CONTINUITY:
Scene layout fixed from frame 1, nothing appears from nowhere. Damage stays. Heroine stays frame-left, enemies frame-right (180-degree rule). Her face or 3/4 profile and the target visible in every strike, never filmed from directly behind. Every strike touches a named point with damage exactly there. Colossus keeps one design and 6x scale. Heroine stays on the ground. One action at a time.
NEGATIVE CONSTRAINTS:
no looking at camera during attacks, no air swings, no objects from nowhere, no frozen background, no direction flips, no damage resetting, no monster design or scale changes, no text or logos, no blood except the nose drop, no wounded or impaled bodies, no flying heroine, no anime look, no magic circles, no runes.
Beyond Scale retweeted
The scariest thing on this rooftop is wearing a blue dress
Created by Seedance 2.5 on @makeugc
Watch her hands at 0:09, not the monsters. Almost nobody catches it the first time.
Prompt:👇
FORMAT: 30 seconds, 16:9, live-action supernatural action thriller.
CINEMATIC QUALITY (applies to every shot):
Must look like real footage from a top-budget live-action feature film, indistinguishable from a real camera shoot, never like AI. Shot on ARRI Alexa 65 with sharp high-end prime lenses, cinematography in the style of Greig Fraser: motivated natural light, soft realistic shadows, true-to-life exposure, deep but detailed blacks, no blown highlights. 24 fps with natural 180-degree shutter motion blur. Native high resolution with razor-sharp focus on the subject, crisp micro-contrast and high-frequency detail on every edge and surface. Subtle organic 35mm film grain over the whole image. Natural, slightly muted real-world color grade with accurate skin tones.
TEXTURE REALISM:
Every surface has real physical texture: visible skin pores, fine facial hair, natural skin imperfections and freckles, individual hair strands, fabric weave and seams, scratches, dust, dirt, wear and moisture on materials, real reflections on wet or polished surfaces. Water lies on surfaces as droplets, puddles and wet sheen rather than as haze in the air. Atmosphere stays clear enough that background detail remains sharp.
MOTION REALISM:
All movement looks performed by real actors and stunt performers: real weight, inertia, momentum, balance and recoil, feet planted on the ground, natural acceleration and deceleration. Every physical contact is visible and has an immediate physical reaction. Camera movement is operated by a real crew: steady, motivated by the action, with a short natural jolt only on heavy impacts.
ANTI-AI RULES:
No motion smoothing, no frame interpolation, no beauty filter, no skin retouching, no waxy or plastic skin, no airbrushed surfaces, no over-sharpened halos, no smeared or melting details in fast motion, no floaty weightless movement, no morphing faces or objects, no oversaturated colors, no fake glow or bloom, no video game or CGI render look, no low resolution, no compression blur.
LIGHT EFFECTS EXCEPTION: The no-glow, no-bloom, no-oversaturation and muted-grade rules apply to everything except the story's own light effects. Power beams, the glowing suit, energy and explosions may glow intensely in their own colors and must behave like real light sources: they cast colored light, reflections and moving shadows on faces, debris and surroundings, and never look like flat overlays.
ACTION ATMOSPHERE: Smoke, dust, ash, sparks and debris produced by the action are allowed and physically driven, staying local to their sources; the main action and background events stay clearly readable.
REFERENCES:
Image 1: Yuna, defines face, hair and outfit. Image 2: hound design. Image 3: alpha design.
CHARACTERS:
YUNA: Korean woman, 19, slim, 165 cm. Long straight glossy black hair below the shoulder blades with soft curtain bangs. Delicate innocent face, natural makeup. Cobalt-blue satin button-up shirt dress above the knee with a thin matching belt, white crew socks, black leather loafers. Same face, hair and outfit in every shot; satin gets wet and wrinkled, never changes color.
HOUNDS: lean humanoid predators running on all fours, 1.5x Yuna's height when standing. Grey-white skin like wet cracked clay, protruding ribs, long arms with black claws, eyeless smooth head with a wide jagged mouth. Fast, low, pack movement, rasping breath, claws scraping concrete. Damage: when hit by fire their clay skin cracks, chars black and breaks apart into chunks and dust. No blood.
ALPHA: the pack leader, same design as the hounds but massive: 4x Yuna's height, heavy shoulders, thick forearms, slow to accelerate, devastating swings. Weak point: a deep crack across the center of its chest. One design and one scale for the whole clip.
ENVIRONMENT + SCENE LAYOUT:
Overcast daytime Seoul after rain, flat grey sky, wet concrete rooftops near the river, distant skyline with a tower on a green mountain and a static yellow tower crane far right. ROOFTOP 1 (start): wet concrete, rusty AC units along the LEFT edge. A 5 m gap with a 40-floor drop separates it from ROOFTOP 2, 2 m lower: yellow-painted helipad circle in the center, a cylindrical steel water tank on its RIGHT side, an empty steel billboard frame (no panel, no text) on its far RIGHT edge, metal railing and a concrete roof edge on its LEFT side. A taller tower stands LEFT of rooftop 2; on its roof edge the ALPHA crouches, still and small in the distance, from the first frame. SIX HOUNDS start chasing on rooftop 1, 4-8 m behind Yuna. These are all the creatures in the film.
POWERS & ENEMY:
FOXFIRE (Yuna's power, only after the reveal): her irises turn glowing amber-gold with vertical slit pupils. Level 1: both palms ignite with dense golden fire, strikes blast sparks and heat, the target cracks and chars at the contact point and is thrown away. Level 2: nine translucent tails of golden foxfire unfurl from behind her lower back, each 3 m long, moving like heavy whips. Recoil: hair and dress hem blown back, puddles under her flash to steam. Cost: breathing hard, tails fade into sparks after use. Sound: deep fire roar, whip crack, heavy impact thud.
Level 1 from 11s, Level 2 only at 23s. Before 9s she shows no power and looks like a helpless victim.
TIMELINE:
0-3s HEROINE: runs away from camera across rooftop 1, center frame, glances back over her left shoulder, terrified. ENEMY: one hound leaps over the camera and lands frame-left 2 m from her, she swerves right; five hounds sprint behind. BACKGROUND ACTION: a hound knocks an AC unit off the left edge, sparks. CAMERA: handheld, running behind and slightly left of her, low. END STATE: Yuna at the edge, leading hound 3 m behind her left.
3-6s HEROINE: plants right foot on the edge and jumps the 5 m gap, arms forward. ENEMY: the leading hound jumps after her, catches her mid-air, claws grip her right shoulder, jaws snap 10 cm from her face; she screams and turns her face away. They land together on rooftop 2; Yuna rolls out to frame-right, the hound slams head-first into the water tank. FX & DEBRIS: the tank's lower front tears, water floods across the helipad. CAMERA: wide low side angle, the drop and skyline visible, jolt on landing. END STATE: Yuna in water near helipad center, hound stunned at the tank.
6-9s HEROINE: crawls backward through the water to the center of the helipad circle, eyes darting. ENEMY: the other five hounds leap the gap and close a half circle behind her; the stunned hound gets up. BACKGROUND ACTION: on the taller tower left the alpha rises and looks down. CAMERA: medium 3/4 front, slow push-in. END STATE: Yuna on her knees facing camera, hands trembling at her chest, six hounds crouched 5 m behind her.
9-11s HEROINE: trembling stops, she lifts her head, slow calm smile, irises flare glowing amber-gold with slit pupils, she looks straight into the lens (no attack in this moment). ENEMY: the hound at her left and the hound at her right launch into the air at her. CAMERA: close-up, hounds blurred behind her. END STATE: two hounds mid-leap from both sides.
11-14s HEROINE: rises and swings both arms out to the sides, palms igniting with golden fire, head turned to the right hound. Right palm strikes the right hound's chest, left palm strikes the left hound's chest. ENEMY: both chests crack and char at the contact points; the right hound slams into the billboard frame and bends it, the left hound flies over the left railing. FX & DEBRIS: golden sparks, clay chunks, steam. CAMERA: wide low front angle, both contacts visible, jolt. END STATE: Yuna standing, arms spread, palms burning.
14-17s HEROINE: a third hound charges from frame-left front; she pivots on her left foot, eyes on it, spinning right roundhouse kick with a trail of golden fire, loafer hits the side of its head. ENEMY: the head cracks, the hound flips across the helipad and breaks into chunks. CAMERA: medium 3/4 front following the kick. END STATE: Yuna landed in a fighting stance, facing frame-left.
17-20s ENEMY: the alpha leaps from the taller tower and lands on rooftop 2's LEFT edge 8 m from Yuna; the concrete edge caves in and cracks. The last three hounds crouch behind its legs. HEROINE: the shockwave of water spray knocks her down to one knee; she looks up frame-left at the alpha, palms smoldering. CAMERA: low angle on the alpha with Yuna in frame for scale, heavy jolt. END STATE: alpha on the cracked left edge, Yuna kneeling right of center.
20-23s HEROINE: rises, raises both burning palms in a cross block, looking at the alpha. ENEMY: the alpha steps in and swings its right forearm horizontally into her crossed forearms. Her block breaks, she is thrown 6 m to frame-right, her back hits the bent billboard frame which bends further; the fire on her palms goes out. CAMERA: low wide side angle, heavy jolt. END STATE: Yuna slumped against the frame, breathing hard; the alpha stomping toward her.
23-26s HEROINE: pushes off the frame, plants her feet, eyes blaze, nine golden foxfire tails unfurl from behind her lower back, puddles flash to steam. ENEMY: the alpha lunges, jaws open. All nine tails whip forward together into one point: the crack in the center of its chest. Golden light bursts through every crack of its body, it is blasted backward and breaks into large charred chunks that fly over the left edge. CAMERA: low 3/4 front, her profile and its chest visible, short slow motion at contact, then real speed. END STATE: Yuna standing, tails behind her, alpha gone.
26-30s HEROINE: turns her head left to the three hounds fleeing toward the cracked left edge; one tail lashes down onto that edge. ENEMY: the edge collapses and drops with the three hounds. Then the tails fade into drifting sparks; Yuna stands in steam among debris, brushes wet hair from her face, eyes fade back to dark brown, she looks into the lens with an innocent face. CAMERA: medium front, slow handheld pull-back to a wide shot with the skyline. Clean final frame.
CAMERA:
Handheld documentary-style operator running with the action, wide-angle prime close to the subject, low angles, real motion blur on fast moves, stable framing on every contact. Maximum 9 camera setups, one per beat, cuts only on motion. Cold overcast daylight look, grey-blue concrete, the cobalt dress and golden fire are the only saturated colors in frame.
AUDIO & DIALOGUE:
Part 1: heavy breathing, running footsteps on wet concrete, claws scraping, rasping snarls, AC crash, Yuna's short scream "No!", tank tearing, water gushing. Reveal: sudden silence, a soft playful exhale. Fight: fire ignition roar, cracking impacts, metal bending, kick whoosh, the alpha's massive landing boom, her pained gasp on the block, deep roar of nine tails, whip cracks merging into one boom, concrete breaking away. End: dripping water, her slow breath, distant wind. No music.
CONTINUITY & STAGING RULES:
- SCENE LAYOUT is fixed from frame 1. Every vehicle, creature and prop that appears in this clip is visible or clearly placed from the first second. Nothing new enters the scene unless the timeline shows it entering from a visible point (street end, alley, rooftop).
- No object changes type, position or state unless the timeline causes it. A parked vehicle stays parked in the same spot until it is hit. Damage stays.
- Maximum 9 camera setups in the clip, one per beat. Prefer one continuous tracking shot. No random angle cuts.
- Camera stays on one side of the action (180-degree rule). Heroine stays center and frame-right, enemy stays frame-left unless the timeline moves them.
- Heroine's face or 3/4 profile and the target are both visible during every strike. No attacks filmed from directly behind her.
- Every strike shows the weapon or power physically touching a named point on the target, with damage appearing exactly at that point. No air swings, no target breaking on its own.
- Monster keeps one design, one head and the same scale (hounds 1.5x, alpha 4x heroine height) for the whole clip.
- Heroine stays on the ground unless a jump is described with take-off and landing.
- No text, letters or logos on vehicles, signs or screens.
- One action at a time. Each beat finishes before the next begins.
NEGATIVE CONSTRAINTS:
No looking at camera during attacks, no attacks on unseen targets, no attacks filmed from directly behind, no air swings, no objects appearing from nowhere, no static or frozen background characters, no idle crowds, no direction flips between shots, no background objects resetting after damage, no monster design or scale changes, no text or lettering on vehicles and signs. No blood, no gore, no neon, no night, no sunshine, no extra creatures beyond those listed, no duplicated hounds, no extra limbs, no music, no subtitles.
🤝 Paid partnership · 🤖 Made with AI
Beyond Scale retweeted
The people sitting beside you deserve your attention more than the strangers waiting behind your screen 🤝
Created this with @makeugc 🎬
🤝 Paid partnership · 🤖 Made with AI
Your AI UGC ads look and sound fake. But they don’t have to.
We just dropped a full video showing how to create AI UGC videos that actually look realistic to use for your ad campaigns.
RT + comment and we’ll send you 300 free credits 👇
youtu.be/5Kd56IFo4o8?si=usiO…
Beyond Scale retweeted
She escaped a lab that ran experiments on people with superpowers.
Her name is Lilit. And she had no idea what was waiting for her outside.
Created with Seedance 2.5 on @makeugc
The world didn't end slowly. A nuclear crisis wiped out the cities in days, and what was left of them filled up with the dead. Streets, subway tunnels, malls. Hordes everywhere.
Almost nobody survived.
The few who did had one thing in common. Something in them changed. Some heal in hours. Some hear a heartbeat through a concrete wall. Some just don't get bitten.
The government noticed.
Instead of rescuing them, it started hunting them. Survivors were tracked down, sedated and taken to an underground facility where they became test subjects. Blood draws, stress tests, exposure to infected tissue. Nobody who went in ever came out.
Drones, patrols, fake rescue broadcasts. Whoever answered the signal got a needle in the neck and a number instead of a name.
Lilit was number 41.
She spent months strapped to a table while they tested how much her body could take. Then one night the backup generator died for 90 seconds. That was enough.
She's out now. No weapon, no map, no idea how far the horde has spread since they took her.
And the people who ran that lab know exactly what she can do. They want her back.
🤝 Paid partnership · 🤖 Made with AI
Beyond Scale retweeted
Made with Seedance 2.5 on @makeugc (native 15-second video, no editing)
PROMPT ⬇️
**CAMERA:**
DV 16mm tape camcorder handheld feel. A self-filmed gym vlog: MILA holds the camera for the opening selfie, then props it securely on a low bench for the jump-rope shots. Slight hand shake, misaligned framing, delayed autofocus and a briefly cropped forehead in the selfie. Once propped, the camera stays still; tighter inserts are simple crops of that same recording. Casual hard cuts omit setup time. The camcorder never appears on screen or in the mirror.
**LOOK:**
Soft, slightly blurry tape quality, faint tape noise, bloomed highlights under gym lighting, subtle auto-exposure fluctuations, muted contrast, realistic skin tones. MILA looks like a real woman recorded in an ordinary gym: natural skin texture, small imperfections, fine loose hairs and spontaneous facial expressions remain visible beneath the gentle tape softness. Soft overhead lighting creates believable shadows beneath her chin and feet. No beauty-filter smoothing or glossy commercial grading.
**STYLE:**
Playful, self-deprecating gym-vlog tone — initial confidence, a few genuinely good jumps, one tiny mistake and an extremely well-earned water break. Real effort mixed with humor. Short, casual cuts, normal-speed movement and a light, funny finish.
**Character**
MILA — a beautiful American Instagram-style brunette model and performer, 20. Long dark-brown hair in a high ponytail, glowing skin with a light sweat sheen, glossy lips, expressive eyes, soft glam makeup and delicate sparkly stud earrings. Slim athletic figure with a very large, full bust, very full rounded buttocks, a narrow waist, flat stomach and lean arms and legs.
She wears a cobalt-blue supportive sports bra with thin straps and a deep scoop neckline showing cleavage, paired with matching ultra-short fitted athletic shorts. Bare shoulders, arms and midriff. White ankle socks, sneakers off. She speaks with an American accent.
Her ponytail follows each jump with a slight natural delay. Her shoulders stay relaxed, knees soften on landing, and her fitted sportswear moves naturally with her body.
**Setting**
A gym mat area in the evening, with a flat rubber exercise floor, a mirror wall off to one side and other equipment visible in the background. In shots 2–4, the camera looks diagonally toward the mirror so MILA's full-length reflection is clearly visible beside her, matching her movements and the rope on every turn, including the criss-cross jump. Soft overhead lighting. Across the aisle, one gym member pedals a stationary bike and another does seated cable rows. Beside the bench, well outside the rope's path, stands a half-full, store-bought 500 ml bottle of water. It has thin clear ribbed plastic, a narrow neck and a small white screw cap. One ordinary dark jump rope with two black handles. Enough clear space above and around her for the full rope arc.
**Storyboard (15s, 6 shots)**
1. *(0–2s, handheld arm's-length selfie)*
MILA looks into the lens with an overly confident little smile, holding both rope handles together in her free hand. One loose strand of hair rests against her cheek. The framing dips slightly as she starts lowering the camera toward the bench.
MILA: “This looks easy.”
2. *(2–5s, propped camera, full-body shot)*
Hard cut: the camera is now resting on the bench, and MILA is already jumping a few steps away. Her whole body, both handles and the rope's overhead arc fit in frame. She makes several small, steady two-foot hops, turning the rope with compact wrist movements, elbows close to her sides. The bottle stands at the edge of frame. Several metres behind MILA, both gym members are visible at their machines, each absorbed in their own workout. Her ponytail bobs behind her.
No dialogue — rhythmic rope taps, soft landings and light breathing.
3. *(5–7s, slightly tighter full-body crop from the same propped recording)*
On one turn, as the rope swings down in front of her, MILA crosses her forearms low at her waist, right wrist over left. She clears the crossed loop with one small two-foot hop. She opens her arms on the next turn and clears that one cleanly too. Both wrists and both feet stay in frame. A quick, pleased grin.
No dialogue. Two evenly spaced taps of the rope on the rubber floor.
4. *(7–9s, back to the full-body framing)*
The rope catches lightly against her toes on the next pass. Its rotation stops; she settles onto both feet and takes one small recovery step, still holding a handle in each hand. She looks down at the slack rope, then back toward the lens with an amused, mildly offended expression.
MILA, with a breathy little laugh: “Okay. Rude.”
5. *(9–11.5s, propped camera, medium crop near the bench)*
Jump cut: she is crouched beside the bench. She places both handles and the slack rope on the floor, then picks up the same water bottle with her left hand and twists off its loosely fitted cap with her right. The cap stays between her fingers. Her shoulders rise and fall with slightly quicker breathing; she smiles to herself.
No dialogue. The thin plastic crinkles softly in her grip over the hum of the exercise bike.
6. *(11.5–15s, closer crop, camera still propped)*
Still crouching near the lens, she brings the bottle's narrow open neck to her lips, tips it for one small sip, lowers it and swallows. The water tilts naturally inside the bottle. She lets out a small satisfied breath, looks straight into the camera and gives a cheeky half-smile.
MILA: “Still counts.”
Finish on her amused expression, with the bottle lowered in her hand.
**Audio and continuity**
Natural American-accented speech with accurate lip sync. Soft gym room tone, rope swishes and floor taps synchronized with each rotation, light breathing, the soft crinkle of the water bottle and a quiet sip. Leave space between actions and dialogue. No music, voiceover, captions or added comedy effects.
Keep MILA's face, proportions, outfit, hairstyle, lighting and surroundings consistent. The rope remains one continuous cord attached to the same two handles: one rotation per hop, no extra loops, disappearing sections or passage through her body. Both hands hold their respective handles throughout the jumping. The same bottle stays beside the bench until she visibly picks it up.
#AIVideo
🤖 Made with AI
Seedance 2.5 vs Kling 4.0. Here’s what a 16× price difference looks like
Same prompt, side by side on @makeugc
1. Seedance got it right on the first try. I generated a second version just to compare and still liked the first one better.
2. I ran Kling twice too. Again, I preferred the first take. So you’re seeing the first result from each model.
3. My first comparison was close. This time, Seedance is way ahead on realism and overall quality. Kling followed the prompt pretty well, though it missed a few details.
4. Kling cost about 1/16 as much per generation. Even at that price, I wouldn’t recommend it for this particular shot. The difference on screen is too noticeable for me.
This was pure text to video. I didn’t use any image or video references.
Would you go with Seedance or Kling for your own videos?
PROMPT ⬇️
**CAMERA:**
DV 16mm tape camcorder handheld feel. MILA films the opening and closing selfies herself; her gym friend films the dumbbell exercises handheld. Hand shake, misaligned framing, delayed focus pulls, clumsy zooms, occasional face cut-off framing, imperfect shots. Camcorder and camera operator never appear on screen.
**LOOK:**
Soft, slightly blurry tape quality, faint tape noise, bloomed highlights under gym lighting, flickering auto-exposure, muted contrast, realistic skin tones.
**STYLE:**
Playful, self-deprecating gym-vlog tone — genuine effort mixed with humor. Quick, casual cuts between excerpts from an ongoing workout. Exercise shots begin and end mid-set, with continuous rhythmic repetitions at normal speed. She talks while moving, without stopping to pose. Time passes between clips; equipment setup and camera handoffs happen off-screen.
**Character**
MILA — a beautiful American Instagram-style brunette model and performer, 20. Long dark-brown hair in a high ponytail, natural skin texture with a light sweat sheen, glossy lips, expressive eyes, soft glam makeup and delicate sparkly stud earrings. Slim athletic figure with a very large, full bust, very full rounded buttocks, a narrow waist, flat stomach, lean arms and slender legs.
She wears a cobalt-blue supportive sports bra with thin straps and a deep scoop neckline showing cleavage, paired with matching ultra-short fitted athletic shorts. Bare shoulders, arms and midriff. White ankle socks, sneakers off. She speaks with an American accent.
**Setting**
A gym dumbbell area in the evening — weight racks, benches, soft overhead lighting. Two other gym-goers exercise casually in the background, away from MILA. An angled mirror naturally reflects her back, blue shorts and legs in wider shots. She uses the same matching pair of small black hex dumbbells throughout.
**Storyboard (15s, 6 shots)**
1. *(0–2.5s, moving arm’s-length selfie)* Already walking past the weight rack, MILA smiles toward the lens and briefly glances at the equipment. Her face and blue sports bra are visible; the framing bobs with her footsteps. MILA: “Arm day. This should be quick.”
2. *(2.5–5s, handheld medium shot)* Cut into an ongoing set of standing alternating dumbbell curls. One arm rises as the other lowers; she keeps the rhythm while speaking. The camera drifts slightly with the operator’s movement. MILA: “These felt lighter yesterday.”
3. *(5–7.5s, handheld wider three-quarter shot)* Another excerpt: MILA is already doing lateral raises, repeatedly lifting and lowering both dumbbells without holding a pose. Her rear view is visible in the mirror. MILA (breathless): “I picked the light ones!”
4. *(7.5–9.5s, handheld close-up insert)* Her hand and forearm grip a dumbbell through the continuing repetitions. The camera briefly loses focus as the weight moves, then catches it again. No dialogue — breathing and distant gym sounds.
5. *(9.5–12s, handheld medium-wide shot)* MILA walks briskly past the camera toward the rack, carrying both dumbbells beside her thighs. The operator turns to follow her as she laughs. MILA: “Yep — my arms just quit.”
6. *(12–15s, arm’s-length selfie after the set)* A clear jump cut: the dumbbells have been returned, and MILA is already holding the camera herself. Walking out of the weights area, she shakes out her free hand and grins. MILA: “Same time tomorrow? No promises.”
#AIVideo
Beyond Scale retweeted
He picked the wrong opponent
Created by Seedance 2.5 on @makeugc
🤝 Paid partnership · 🤖 Made with AI



















