A creator who says your line, in a kitchen that looks like a kitchen, shot like a phone. The face, the room, the delivery and the voice all arrive in the same pass — there is no separate actor to book, no voice to clone and no lip-sync step to fix afterwards. Your first one is free. No card.
Five creator-style ads and the complete prompt behind each. Every card opens the whole thing.
The whole prompt is a look
The script on the left. What came back on the right.
Five UGC ads with the whole prompt printed, not a summary
The whole prompt is a look16:9 · 15s · text only, no reference
The shortest prompt here and the only one with no shot list at all — one long stack of adjectives about a look. It works, and it costs you the pacing: the model chose where the cuts go, and you cannot ask for that arrangement again. Useful when you want a mood board that moves. Not useful when the brief says the hook lands by second two.
UGC-style skincare video featuring a realistic young woman speaking directly to the camera in front of a bathroom mirror while holding a premium facial serum bottle. She smiles naturally, talks about her skincare routine, and demonstrates applying the serum to her face. Authentic, conversational presentation with genuine facial expressions and hand gestures, natural morning light, casual home bathroom setting, clean skincare aesthetic, healthy glowing skin, minimal makeup, smartphone-shot look, candid lifestyle beauty content, soft depth of field, luxury beauty brand feel, vertical 9:16, social media ad quality, highly realistic natural skin texture, not overly retouched.
Nine beats with a clock on them16:9 · 15s · text only — it asks for two images it never got
Nine beats, each with a start and an end and the line she says in it. This is the shape to copy. Two things worth noticing before you do: it opens by asking for 9:16 vertical format and came back wide, because the shape is a control and not a sentence — and it says use her facial features exactly as shown in her reference photo when no photo was attached, so the model simply invented a creator. She holds across all fifteen seconds. She will not hold across tomorrow’s ad.
CHARACTER & WORLD LOCK + FULL 15-SEC MASTER PROMPT
A 15-second continuous UGC-style vlog in 9:16 vertical format. Jessy, a woman in her mid-20s — use her facial features exactly as shown in her reference photo, keep everything else about the scene freshly designed — wears an attractive pastel-colored dress (soft lavender, mint, or buttery yellow) that visually pops in contrast against the bubblegum-pink Olipop can. Location: a bright, clean modern kitchen/countertop with soft natural daylight. Product: a can of Olipop Blackberry Vanilla — pastel pink can, magenta blackberry illustration, white vanilla flower, dark green "Blackberry Vanilla" wordmark, "Supports Digestive Health" text near the top, "A New Kind of Soda™" tagline (use the uploaded product photo as the exact visual reference for the can). Style: authentic smartphone UGC vlog, natural handheld feel, realistic skin texture, no over-retouched look. She speaks casual, upbeat, conversational English directly to camera with natural lip-sync.
The video follows this exact beat sheet, with clean cuts between each moment:
00:00–00:02 — Front-camera selfie, slight handheld sway, Jessy fans herself lightly in the kitchen, looking for something refreshing, smiles at camera: "Okay I need something fun to drink right now."
00:02–00:04 — Eye-level medium shot, she opens the fridge and pulls out the pink Olipop can, face lighting up with delight: "Ooh, Blackberry Vanilla Olipop — yes please."
00:04–00:06 — Close-up on her hands slowly turning the can, label details ("Prebiotics, Botanics, Plant Fiber") visible: "Prebiotics AND it tastes like dessert? Sold."
00:06–00:08 — Tight macro shot, her fingers pop the tab, condensation and a light fizz mist visible: "Listen to that fizz."
00:08–00:10 — Slow-motion macro shot, pink soda pouring over ice into a clear glass, bubbles rising: "Look how pretty that color is."
00:10–00:12 — Close-up on her face as she sips, eyes lighting up with genuine, pleasantly surprised reaction: "Wait... that's actually so good."
00:12–00:13 — Macro hero shot of the can alone on the counter, condensation dripping, soft bokeh kitchen background, slow subtle push-in, no dialog.
00:13–00:14 — Medium shot, she smiles at camera, can raised slightly at chest height: "Blackberry vanilla was the right call."
00:14–00:15 — Close front-camera selfie, can held beside her face, warm bright smile into the lens: "Go grab an Olipop, seriously."
Continuous natural ambient kitchen sound, subtle upbeat pop background music underneath, her voice clear and casual throughout. Consistent face, outfit, and setting across every beat. No text overlays, no captions, no watermarks, no horizontal framing, no changes to the can's design or branding.
Eight numbered shots in fifteen seconds, which is a cut every 1.9 seconds, plus a named product image the pizza has to stay faithful to. Three short lines of dialogue and no voiceover. The end card did not survive: shot 8 asks for a logo close-up with pizza night = solved set over it and a cut to black, and what comes back is still on her face. Type in the last second is the beat to composite rather than to generate.
VIDEO PROMPT — "Pizza Night Vlog" (UGC iPhone Style)
Duration: 15 seconds | Aspect Ratio: 16:9 | Style: Authentic UGC / iPhone selfie-vlog, handheld, natural light, slight motion blur, TikTok/Reels energy — NOT cinematic, NOT overly polished. Feels like a real creator filmed this on their phone.
Product Reference: Use the uploaded Domino's Pepperoni Pizza image as the only product reference. Keep crust thickness, cheese texture, bake color, pepperoni placement, and proportions identical in every cut — no redesigning the pizza.
Camera: iPhone 15 Pro front + back camera switching, handheld, natural wobble, autofocus hunting slightly (realistic), vertical-style framing cropped to 16:9, occasional finger near lens edge, natural room lighting + phone flash reflections on the pizza box.
Character Description
Name (for reference): Mia
Awoman in her mid-20s, naturally attractive and beautiful with an approachable, girl-next-door charm — not overly done up. Wavy sandy-blonde hair pulled back loosely, light natural makeup, wearing a cozy oversized cream sweater. Warm, genuine smile, expressive eyes, casual energetic personality like a real lifestyle vlogger. Sits in a softly lit modern kitchen/living room.
Shot Breakdown
SHOT 1 (0–2s) — The Grab
Selfie-angle, she's mid-laugh holding up the Domino's box to camera. Quick jump cut. Dialogue: "Okay so it's officially pizza night—"
SHOT 2 (2–4s) — The Open
Cut to overhead handheld shot, box flips open, steam rising off the pizza, slight camera shake as she leans in.
SHOT 3 (4–6s) — The Zoom
Quick zoom-punch into the pizza, phone camera autofocus adjusts naturally, cheese and pepperoni in focus, ambient kitchen sounds.
SHOT 4 (6–8s) — The Pull
Cut to her hands lifting a slice, natural cheese pull, filmed from a slightly low candid angle like a friend filming across the table.
SHOT 5 (8–10s) — The Reaction
Cut back to selfie-cam, she takes a bite, eyes widen, quick genuine reaction. Dialogue: "Oh my god, that's so good."
SHOT 6 (10–12s) — The Candid Cutaway
Jump cut to a close, slightly shaky shot of the pizza box on the counter, her hand grabbing another slice off-frame, casual b-roll energy.
SHOT 7 (12–14s) — The Wrap-Up
Back to selfie angle, she grins at camera, holding slice up like a toast. Dialogue: "Dominos, y'all know what to do."
SHOT 8 (14–15s) — End Tag
Quick freeze/cut to the box logo close-up, natural handheld wobble, soft text overlay in casual font: "pizza night = solved 🍕" — cut to black.
Look & Feel
Warm indoor lighting, slightly grainy natural phone sensor look, imperfect framing, real reactions, minimal dialogue (3 short lines total), authentic pacing with hard jump cuts instead of smooth transitions.
Negative Prompt
cinematic grade, overly smooth camera moves, studio lighting, professional voiceover, staged acting, CGI look, plastic cheese, distorted pepperoni, extra fingers, warped hands, text glitches, logo distortion, overly polished commercial feel.
One creator you can cast again16:9 · 15s · 1 reference image, identity locked
The only one here that attaches a face, and the reason the page bothers to separate them. The opening sentence — Preserve her facial identity, hairstyle, eye color, skin tone, body proportions — is not what holds her. The uploaded still is. Run this prompt again with the same image and you get the same creator; run the soda prompt again and you get a stranger. That is the whole difference between one ad and an ad account.
Use the uploaded reference image as the exact character reference. Preserve her facial identity, hairstyle, eye color, skin tone, body proportions, and facial consistency throughout every shot.
Create a 15-second ultra-realistic UGC-style beauty advertisement featuring a realistic young woman speaking directly to the smartphone camera in a bright, modern bedroom vanity area with warm morning sunlight, soft neutral décor, and an elegant illuminated mirror.
She smiles naturally while holding a premium luxury lip oil with a clear glass bottle and gold cap. In a friendly, conversational tone, she shares why it has become her favorite everyday beauty essential. She opens the product, applies the lip oil with the applicator in front of the mirror, then gently presses her lips together and smiles confidently, showing the glossy natural finish.
Capture authentic handheld selfie shots, over-the-shoulder mirror angles, and cinematic close-ups highlighting the crystal-clear bottle, gold cap, soft reflections, applicator, glossy texture, and luxurious packaging. End with her holding the lip oil beside her face while smiling naturally into the camera.
Authentic UGC creator content, realistic facial expressions, natural hand gestures, healthy glowing skin, minimal makeup, luxury beauty aesthetic, smartphone-shot realism, subtle depth of field, premium color grading, vertical 9:16, 4K HDR, no text, subtitles, logos, or watermarks.
The look is a camera, not a grade16:9 · 15s · text only
The look is specified as a camera rather than a grade: DV 16mm tape camcorder, propped on a gym bag, delayed focus, clumsy zooms, face occasionally cut off. Every one of those is a thing a lens does, which is why they survive the render — where authentic and raw do not. Six cuts, five spoken lines, no music named.
CAMERA: DV 16mm tape camcorder POV. CHASE films herself, occasionally propping the camera on a wall or gym bag. Natural handheld shake, imperfect framing, delayed focus, clumsy zooms, face occasionally cut off. Camcorder never visible.
LOOK: Soft blurry tape aesthetic, faint tape noise, bloomed gym lights, flickering auto-exposure, muted contrast, realistic skin.
STYLE: Calm post-workout gym vlog. Relaxed, reflective, slightly tired but happy. Slow pacing, natural laughs.
CHARACTER: CHASE, Korean idol (20s). Long black hair loosely tied, light post-workout glow, slim athletic build. Modest long-sleeve athletic top, loose joggers/leggings, socks or shoes off, towel around neck, no jewelry.
SETTING: Quiet evening gym. Stretching mat, mirror wall, gym bag, protein shake, soft lighting, mostly empty.
15s | 6 Cuts
Propped medium: sits on mat, exhales. "Okay, cool-down time."
Handheld stretch: hamstring stretch, laughs. "Yeah, that was a lot today."
Macro: picks up protein shake, condensation visible. Ambient sound only.
Medium: drinks shake, smiles. "Okay, that's actually really good."
Close: overhead stretch, rates workout. "I'd give today like an eight out of ten."
Selfie: grabs bag and shake, waves. "Alright, I'm heading out—see you guys."
What this is, and what it is not
An ad you can throw away. That is the whole point.
Nobody needs one UGC ad. A test needs six hooks against the same offer, and the hook that wins is the one you could not have guessed. The reason that loop is expensive has never been the edit — it is that every variant costs an actor another take. So the question is not whether a generated creator beats a real one. It is what a variant costs, and how many you can be wrong about before it stops being worth it.
The category prices per seat and per video, and mostly does not publish it. Arcads lists $110 a month for ten videos — about $11 each — and its own pricing page returns a 404, so you see the number after you have signed up. Creatify runs $39 a month for roughly 100 credits and $99 for 300, which lands somewhere near $1.65–$2 a video if you use the whole allowance and rather more if you do not. Everything above that is a production: a UGC creator on a marketplace is $150–$500 a video plus the brief and the wait. Fifteen seconds here is 125 credits, about $2.75, and there is no monthly floor. An eight-second hook is 69 credits. A failed generation is refunded automatically, which is the number that actually matters when you are testing six openings you expect five of to lose.
Arcads Starter — ten videos, ≈$11 each
$110/mo
Arcads Starter — ten videos, ≈$11 each
Creatify, a month, before you render anything
$39–$99
Creatify, a month, before you render anything
A brief to a human UGC creator
$150–$500
A brief to a human UGC creator
Fifteen seconds here, voice included
~$2.75
Fifteen seconds here, voice included
No monthly seat to keep paying between tests
No separate voice clone or lip-sync pass
No watermark on any tier, including the free one
No surcharge on the vertical cut
What you put in
Four ways to cast the ad, depending on how many you need
They all run the same model at the same rate. What changes is how much of the result you can ask for a second time — which is the only question that matters once one hook starts working and you need nine more like it.
01
Write the creator, run once
No upload at all. Describe who is on camera and the model casts somebody — consistently, for the length of that clip. Fastest way to find out whether a hook is worth shooting. Three of the five ads above went this way.
One-off tests · text to video
02
Cast a face and keep her
Attach one clean frontal still and the same creator comes back on every run. This is the difference between an ad and an ad account: fifteen variants that a viewer reads as one person, generated a week apart.
The same creator, again · up to 9 images
03
Hand it the product too
The creator in one image, the pack shot in another. Nine references in a single request is a face plus a colourway run — the same script, the same delivery, a different SKU in her hand each time.
Face + product · 9 images, one rate
04
Borrow an ad whose pacing works
Give it a clip off your own swipe file and it takes the cut points and the camera behaviour from that, not the footage. Useful when you know the winning rhythm and cannot describe it.
Copy the edit · up to 3 clips
All four bill the same: 8 credits a 768P second, 13 at 2K, plus 5 for the clip. The first five reference images are free and the sixth onward is 4 credits each. Every limit and rate, with the date each last changed.
Start to finish
How to make a UGC ad in four steps
Around four minutes, most of it spent deciding what she says. The order below is the order the decisions actually have to be made in — each one is cheap now and expensive after you have rendered nine variants.
01
Cast her before you write a word
If this ad has siblings, attach a face now. A name and a paragraph hold a creator together inside one clip and not across two, so a description you liked in variant one is a stranger in variant four. One clean frontal still, reused on every run, is the whole fix.
One frontal stillReuse it every run
02
Put a clock on every shot
00:00–00:02… 00:02–00:04… 00:04–00:06… Naming the seconds is the single biggest lever on whether the hook lands where you meant it to. Eight shots in fifteen seconds is a cut every 1.9 seconds, and that is the register a feed reads as real.
00:00–00:02A cut every ~2s
Propped medium: sits on the mat, exhales, and says: “Okay, cool-down time.” Ambient gym only, no music.
PictureSound
03
Write the line, not a description of the line
Put the words in quotation marks and the model speaks them — the voice is generated with the picture, so there is no clone to train and nothing to sync afterwards. Budget about three words a second, and say what the room sounds like underneath.
One field, both channels≈3 words a second
Dialogue + room tone · 32 kHzHOOK-01.MP4
04
Test at 768P, master the winner at 2K
A 768P second is 8 credits against 13 at 2K, and upgrading a finished 768P clip afterwards costs exactly what rendering it at 2K would have — so the five hooks you kill cost you less than the one you keep. A failed generation is refunded automatically.
Five things that will cost you a render. None of them are in the marketing.
Every one of these came out of the five ads above rather than off a spec sheet, and every one is checkable against the prompts printed on this page — which is more than the category leaders will give you. We went through 1,596 visible words on the biggest UGC ad generator page on the web: it uses the dollar sign zero times, the word “credit” zero times, the word “actor” zero times, and the word “disclosure” zero times.
01The prompt cannot set the shape
Three of the five prompts above open by asking for 9:16 vertical. All three came back 16:9, because the ratio is a control on the generator and the sentence is just more description. Set it on the chip. It bills the same either way.
ratio is a chip, not a sentence
02A named creator is not a cast creator
Call her Jessy, give her an age, a dress and a kitchen, and she will hold for fifteen seconds. Run it again next week and a different woman answers to the same paragraph. If the account needs one recognisable face across a dozen ads, that face has to be a file you attach.
identity = an image, not a name
03Type in the last second is a coin flip
Four of the community scripts we read end on an on-screen tag — “pizza night = solved”, “burger cravings = solved”. The one we ran did not deliver it; the clip is still on her face at 15s. Generate the performance and composite the end card, which is a thirty-second job in whatever you already edit in.
composite the end card
04Fifteen seconds is the ceiling, not a soft one
There is no thirty-second option and no stitching in the product. A UGC ad that has to run 30s is two generations and a cut, and the two halves will not match unless the same reference image is attached to both.
4–15s, hard
05Land the cut on 8.000 seconds
H3 renders on a 17n+5 frame grid at 24 fps, so most lengths come back a fraction long — 15s is really 15.083s. 192 frames, exactly 8.000 seconds, is the only whole second in the range, which makes it the hook length that cuts to a music bed without drifting.
192 frames = 8.000s
An eight-second hook at 768P is 69 credits, about $1.50. Six of them is a test. On a $110-a-month plan for ten videos, six is most of your allowance.
Why this and not an avatar tool
The voice is not a second product
The usual UGC stack is three purchases: a stock avatar, a voice clone, and a lip-sync pass to marry them. That is why the results all share a particular deadness — the mouth is solving for a waveform that was recorded somewhere else. Here the line is spoken in the same pass that draws the face.
One pass, both channels
She is not dubbed, she is speaking
32 kHz stereo generated alongside the picture: the dialogue, the room tone, the fizz of the can. No voice library to license, no sync step, and the breath before the line lands where a breath goes.
Vertical at one rate
9:16 costs what 16:9 costs
Six shapes — 21:9, 16:9, 4:3, 1:1, 3:4, 9:16 — all billed per second, because the upstream bills per second and ignores the frame. The cut every platform actually wants carries no surcharge.
Twelve files in
A face and a product in the same request
Up to 9 reference images, 3 clips and 3 audio files, 12 files total. That is the creator, the pack shot, the colourway and a swipe-file clip for the pacing, in one call. Most hosted doors onto this model do not publish their cap.
No seat
You pay for seconds, not for a month
There is no monthly floor and nothing expires for a year. The tools in this category start at $39 a month and the actor ones at $110; this is 8 credits a second and a real clip free to start.
What you give up
A real person, and the paperwork that comes with one. A generated creator has no signed release, cannot be booked for the follow-up, and cannot tell you the product actually worked. If the ad turns on a claim only a customer can make, cast a customer. What this replaces is the six variants you would never have paid a human to shoot in order to find out which opening line survives contact with a feed.
What it actually costs
An eight-second hook is 69 credits. About $1.50.
A clip is a flat 5 credits for the prompt-understanding pass, plus 8 credits per 768P second or 13 per 2K second. The first five reference images are free, so a creator still and a pack shot add nothing. One credit is one cent of MiniMax’s own published list price, which means the whole table can be checked against a page this site does not control. A generation that fails is refunded automatically — the number that decides whether a six-hook test is affordable.
A UGC ad is a person appearing to endorse a product, which is the most regulated thing on this site. The category leaders say nothing about any of it — zero mentions of likeness, disclosure or licensing across their whole page. Here is the position before you spend anything.
Commercial useIncluded, on the free clip as well. No watermark on any tier, no per-video royalty and no separate licence to buy. MiniMax claims no rights over the Outputs you generate and this site adds none of its own.
Whose face is itA creator the model invents is nobody, and that is the safe case. Uploading a photograph of a real person to put words in their mouth is not: you need their permission, in writing, for that use — and several jurisdictions treat a synthetic endorsement without it as a distinct offence rather than a licensing dispute.
DisclosureMeta, TikTok and YouTube all now require realistic synthetic media to be labelled, and each has its own switch for it in the upload flow. An ad that reads as a real person filming in a real kitchen is exactly the content those rules were written for. Tick the box.
What she may sayA generated testimonial is not a testimonial. The advertising rules on substantiation do not soften because a model drew the speaker, and “results” a viewer would read as somebody’s real experience need to be somebody’s real experience.
Numbers off real generations rather than a model card — the frame grid a cut has to land on, what a 2K second really costs, and how the sound gets written into the same field as the picture.
The eight questions a media buyer asks before putting a generated creator in front of paid traffic, answered with the numbers behind them.
01
Is this AI UGC ad generator free?
The first clip is, and it is the real model — 4 seconds, 768P, with the voice, no watermark, no card, and it never expires. Every ad on this page plays without an account. After that an 8-second hook is 69 credits and a 15-second ad is 125; credits start at $9.9, which puts a full ad at about $2.75. A generation that fails costs nothing.
02
Do I need a separate voice or lip-sync tool?
No, and that is the main reason to use this rather than an avatar stack. Put the line in quotation marks and it is spoken in the same pass that draws the face — 32 kHz stereo, room tone included. There is no voice to clone, no waveform to upload and no sync step. If what you have is an existing clip and a new audio track, that is a different tool: lip sync.
03
Can I use the same creator across a dozen ads?
Yes, but only by attaching her. Describing a creator holds her together inside one clip and not across two — the soda ad above names her, describes her, and would cast somebody else on a re-run. Attach one clean frontal still to every generation instead. Published measurements put reference-conditioned similarity between two clips at roughly 0.54–0.59, so it is a strong resemblance rather than a lock; consistent character video covers what that actually looks like.
04
Does vertical cost more?
No. 21:9, 16:9, 4:3, 1:1, 3:4 and 9:16 all bill at the same per-second rate, because the upstream charges by the second and ignores the shape. Set it on the ratio chip — writing “9:16” into the prompt does nothing, which three of the five prompts on this page demonstrate.
05
Can I make a 30-second ad?
Not in one generation. The range is 4 to 15 whole seconds at 24 fps, so 30 seconds is two clips and a cut in your editor — and the two halves will only match if the same reference image is attached to both. For a hook test that limit rarely bites; almost nothing in a paid feed survives past second three anyway.
06
Do I have to label it as AI on Meta or TikTok?
Yes. Meta, TikTok and YouTube each require realistic synthetic media to be disclosed and each provides the control in the upload flow. A creator-style ad is the exact case those rules describe. From our side there is no watermark and commercial rights are included on every tier; the labelling obligation is yours and it is not optional.
07
Can I upload a photo of a real person?
Only with their written permission for this use. A model-invented creator is nobody and carries no likeness problem; a real face saying a line they never said is a different thing entirely, and several jurisdictions treat an unauthorised synthetic endorsement as its own offence. If you are working with a real creator, get the release the same way you would for a shoot.
08
How is this different from Arcads or Creatify?
Those sell a library of licensed actor performances and charge for a seat: Arcads lists $110 a month for ten videos — roughly $11 each — and Creatify $39 to $99 a month for an allowance you lose if you do not use it. You get real, released faces and a fixed set of them. Here you write the performance instead of picking it, the voice comes with the picture, and you pay 8 credits a second with no monthly floor. Worse if you want a vetted actor roster; better if you want six hooks tonight for the price of one of theirs.
Six hooks tonight, or one brief and a fortnight.
Write who is on camera, what she says and when she says it, and the ad comes back with her voice already in it. Your first one is free — no card, no watermark, and it is yours to run commercially.
