Kling AI Prompts: 24 Tested Examples and a Prompt Formula
24 Kling AI prompts for Kling 3.0, covering UGC hooks, product shots, multi-shot scenes and dialogue with native audio. Each prompt comes with a note on why it works, next to Kling's own prompt formula, tips and the mistakes that waste credits.
★★★★★ 4.9/5 · 27,000+ users
Updated September 25, 2026.
Prompt structure
The Kling AI prompt formula
Kling's own prompt guide builds a prompt from subject, action, setting, camera language, lighting and mood, and asks you to describe visible motion ('smoke drifting upward', 'camera tracking beside the subject') rather than abstract claims. For multi-shot, label each shot and give it its own framing and camera move. For dialogue, tag each speaker with a tone and language in parentheses so native audio assigns the right voice.
Strengths and limits
What Kling AI is good at
Good at
- Native audio in the same pass: dialogue, sound effects and ambience in Chinese, English, Japanese, Korean and Spanish, including accents such as American, British and Indian English
- Multi-shot storyboards in a single generation, either planned automatically or shot by shot with your own framing, camera move and duration
- Character and product consistency through Elements: upload 2-4 reference images or a short video and bind the subject so it survives camera moves and scene changes
- Legible text in the frame for labels and signage, which matters for product shots
- Flexible length from 3 to 15 seconds per generation
Limits
- Duration 3-15 seconds per generation (VIDEO 3.0 and 3.0 Omni)
- Resolution 720p and 1080p; a native 4K mode is offered at a higher credit cost
- Native audio languages: Chinese, English, Japanese, Korean, Spanish
- Credit cost goes up with native audio and resolution (for example 12 credits/second for 1080p with audio vs 8 without, per Kling's user guide)
- Content moderation applies on the Kling platform; prompts or references that break the platform rules are blocked
Copy and paste
24 Kling AI prompts to copy
UGC ad
1. Mirror try-on haul
A woman in her twenties with curly dark hair stands in front of a bedroom full-length mirror holding her phone, trying on an oversized cream knit cardigan. She turns side to side, pulls the sleeves over her hands and grins at her reflection. Cozy bedroom with fairy lights and a messy bed behind her. Camera: mirror-selfie framing, vertical, slight handheld wobble. Lighting: warm evening lamp light. Mood: giddy, casual. Woman (excited, English): "It's so soft I'm not taking it off. Ever."
Why it works: Mirror-selfie framing and a single speaker label give Kling everything it needs for an authentic try-on clip with native audio.
UGC ad
2. Two friends, one blender
Shot 1: Medium shot in a bright small kitchen. Two roommates, Jo in a yellow t-shirt and Sam in a grey tank top, stand behind a counter with a compact mint-green blender. Jo drops frozen berries in. Steady camera. Shot 2: Close-up of the blender as it crushes ice smoothly in a few seconds. Shot 3: Medium close-up of Sam taking a sip, eyebrows up. Jo (teasing, English): "Told you." Sam (surprised, English): "Okay, that's actually smooth." Sound: blender whir, ice cracking.
Why it works: Custom multi-shot splits the demo into setup, proof and reaction, and labelled speakers keep the lines with the right person.
UGC ad
3. Commuter backpack review
A young man in a navy rain jacket walks out of a subway exit into light rain, wearing a charcoal roll-top backpack. He stops under an awning, swings the bag round and unrolls the top to show a dry laptop inside. City street, wet pavement, taxis passing. Camera: tracking shot beside him at shoulder height, then a steady medium close-up when he stops. Lighting: overcast daylight, cool tones. Mood: practical, confident. Man (dry humour, English): "Four months of this weather. Laptop's never been wet."
Why it works: One tracking move into a static reveal keeps motion readable, and the proof moment is the laptop.
UGC ad
4. Skincare shelfie with voiceover
Close-up of a bathroom shelf with five skincare products. A hand with a gold ring picks up a small amber dropper bottle whose label reads "NIGHT OIL" in white serif letters and shows it to the camera. Camera: slow push-in, centered composition, shallow depth of field. Lighting: soft morning window light, warm highlights. Mood: calm routine. Voiceover (relaxed, British English): "The only step I'd keep if I had to drop everything else."
Why it works: Quoted label text uses Kling 3.0's text rendering, and a British-accent voiceover uses its accent control.
UGC ad
5. Street interview hook
Shot 1: Medium shot on a busy sidewalk. A host in a denim jacket holds a small microphone toward a woman in a red scarf. Handheld, eye level. Host (upbeat, American English): "Guess the price of these." Shot 2: Close-up of the woman holding a pair of white leather sneakers, turning them over. Woman (unsure, English): "Two hundred?" Shot 3: Medium close-up of the host laughing. Host (laughing, English): "Sixty-five." Sound: traffic, footsteps, city chatter.
Why it works: The street-interview format is a staple short-form ad, and three labelled shots map directly onto Kling's custom multi-shot.
product
6. Watch on black water
A steel dive watch with a green bezel rests on a black slate slab covered in a thin film of water. Droplets bead on the crystal. Camera: extreme close-up, slow orbit from the left side to the dial, shallow depth of field, centered. Lighting: a single hard spotlight from above, deep black background, bright specular highlights on the steel. Mood: precise, premium. Sound: a soft ticking, water dripping once.
Why it works: One slow orbit and one hard light source is the kind of controlled camera language Kling follows reliably.
product
7. Canned drink ice pour
A matte coral soda can with the word "FIZZ" in bold white letters stands on a pale blue surface. Ice cubes drop around it and roll to a stop, then a hand tilts the can and pours into a tall glass. Camera: medium close-up, static, rule-of-thirds with the can on the left. Lighting: bright high-key studio light, soft shadows. Mood: fresh, playful. Sound: ice clinking, a crisp can opening, fizzing.
Why it works: Quoted brand text, clean composition and three distinct sounds make a short, loopable drink ad.
product
8. Laptop on a floating desk
A thin silver laptop opens by itself on a minimalist oak desk in a white studio. The screen lights up with a soft gradient. Camera: slow push-in from a medium shot to a close-up on the keyboard, centered. Lighting: soft top light with a warm rim from behind, clean white background. Mood: calm, modern. Sound: a quiet hinge, a gentle startup chime.
Why it works: A single push-in synced to the screen lighting up keeps the product the focus from first frame to last.
Run these prompts without five subscriptions
- Paste the prompt into Veo 3.1, Kling 3.0, Seedance 2.0, Nano Banana Pro and GPT Image 2, all in one place
- Turn the clip into a TikTok, Reel or Short with captions and a hook
- Publish straight to TikTok, Instagram and YouTube
product
9. Sunscreen at the beach club
A white sunscreen tube with a blue cap lies on a striped beach towel next to sunglasses. A hand picks it up and squeezes a line of cream onto the back of the other hand. Camera: overhead close-up, then a quick tilt to the sea behind. Lighting: harsh midday sun, bright blues and whites. Mood: summer, bright. Sound: waves, distant laughter, a cap clicking open.
Why it works: An overhead product moment followed by a tilt to context shows the product and where it is used.
product
10. Espresso machine morning
Shot 1: Close-up of a brushed-steel espresso machine as a portafilter locks in with a firm twist. Steady camera. Shot 2: Extreme close-up of espresso streaming into a small white cup, thick crema forming. Slow motion. Shot 3: Medium shot of a woman in a linen shirt lifting the cup by a sunny kitchen window, smiling. Lighting: warm morning sunlight. Mood: slow, ritual. Sound: grinder buzz, pump hum, the cup set on a saucer.
Why it works: Three shots move from mechanism to result to the person, a standard product-story arc in 10-15 seconds.
social hook
11. Wait for it
A man in a hoodie holds a phone at arm's length in a messy garage, pointing at a folded cardboard box on the floor. Suddenly the box unfolds on its own into a small standing desk. Camera: medium shot, static, slightly low angle. Lighting: flat fluorescent garage light. Mood: surprise, comedic. Sound: cardboard creaking, a satisfying click. Man (amazed, American English): "Wait. Wait for it. That's the desk."
Why it works: An impossible physical beat in the middle of an everyday scene is an effective scroll-stopper.
social hook
12. Three things in my bag
Top-down close-up of a woman's hands emptying a tan leather tote onto a white bedspread. She lines up three items one by one: a lip balm, a compact charger, a tiny notebook. Camera: overhead, static, centered. Lighting: soft daylight. Mood: tidy, satisfying. Sound: items tapping on fabric, a zipper. Woman (voiceover, cheerful, English): "Three things. Every day. No exceptions."
Why it works: Lists with a number promised in the first second hold attention, and the flat-lay framing is easy for Kling to keep stable.
social hook
13. Before and after haircut
Shot 1: Medium close-up of a young man with messy, overgrown hair sitting in a barber chair, looking unsure. Static camera. Shot 2: Quick cut to the same man with a sharp fade haircut, turning his head side to side in the mirror. Slow push-in. Barber (proud, English): "Same guy." Lighting: warm barbershop bulbs, chrome reflections. Mood: transformation. Sound: clippers buzzing, then a cape being pulled off.
Why it works: Transformations work as hooks, and binding the man as an Element keeps his face consistent across the two shots.
social hook
14. POV grocery run
First-person view pushing a shopping cart down a bright supermarket aisle. A hand reaches into frame and grabs a green box of oat granola from a shelf, turns it to show the label, then drops it in the cart. Camera: POV, steady forward movement at walking pace. Lighting: bright store fluorescents. Mood: everyday. Sound: squeaky cart wheel, store music faint in the background.
Why it works: POV framing makes the viewer the shopper, a low-cost way to put a product in context.
talking head
15. Nutritionist at the counter
A nutritionist in her thirties with a neat bun and a sage-green blouse sits at a kitchen island with a bowl of lentils in front of her, speaking directly to the camera. Camera: medium close-up, eye level, static, centered. Lighting: soft window key light from the left, warm fill. Mood: friendly, knowledgeable. Nutritionist (calm, clear, English): "If you only change one thing this week, add a handful of these to lunch." Sound: quiet kitchen ambience.
Why it works: A clear speaker description, static framing and one line is the most reliable setup for a Kling talking head.
talking head
16. Bilingual shop owner
A shop owner in his fifties with a grey beard stands behind the counter of a small hardware store, speaking to camera. Camera: medium shot, eye level, static. Lighting: warm overhead bulbs, shelves of tools softly out of focus. Mood: proud, welcoming. Owner (warm, Spanish): "Llevamos treinta años en este barrio." Owner (warm, English): "Thirty years on this street, and we still cut keys by hand." Sound: a bell over the door, faint radio.
Why it works: Kling 3.0 supports code-switching within one scene, so the same speaker can say a line in two languages.
Run these prompts without five subscriptions
- Paste the prompt into Veo 3.1, Kling 3.0, Seedance 2.0, Nano Banana Pro and GPT Image 2, all in one place
- Turn the clip into a TikTok, Reel or Short with captions and a hook
- Publish straight to TikTok, Instagram and YouTube
talking head
17. Podcast two-shot
Two podcast hosts sit at a round table with microphones in a studio with acoustic panels. Lena, in a black turtleneck, and Marco, in a denim shirt. Multi-shot auto: cover the exchange with a two-shot and alternating close-ups. Lena (curious, English): "So why did you quit the agency?" Marco (laughing, then serious, English): "Because I was selling other people's ideas better than mine." Lighting: warm practical lamps, dark background. Mood: candid.
Why it works: Auto multi-shot handles the coverage while labelled lines keep each voice on the right host.
talking head
18. Real estate agent walk-in
A real estate agent in a light grey suit opens the front door of a bright empty apartment and turns to the camera. Camera: tracking shot following her through the door, then a steady medium shot as she stops in the living room. Lighting: natural daylight through large windows, white walls, oak floors. Mood: upbeat. Agent (enthusiastic, American English): "South-facing, two bedrooms, and you're five minutes from the park." Sound: footsteps on wood, door closing.
Why it works: A walk-and-talk gives a property tour with one tracking move and one line.
cinematic
19. Samurai in the bamboo storm
Shot 1: Establishing wide shot of a bamboo forest in heavy wind at dusk. A lone samurai in dark indigo armour stands in a clearing. Slow push-in. Shot 2: Medium close-up of his hand tightening on the sword hilt. Leaves whip past. Shot 3: Low-angle full body shot as he draws the blade in one quick motion. Lighting: cold blue dusk, a thin line of orange on the horizon. Mood: tense. Sound: wind roaring, bamboo creaking, the ring of a drawn blade.
Why it works: Each shot has one camera move and a clear shot size, the approach Kling's storyboard feature is built for.
cinematic
20. Rooftop confession
Two friends sit on a city rooftop at night, legs over the edge, city lights below. Ava, in a red jacket, and Theo, in a grey hoodie. Camera: slow arc from behind them to a side profile two-shot. Lighting: cool blue city glow, warm light from a window behind. Mood: quiet, emotional. Ava (soft, hesitant, English): "I'm moving next month." Theo (after a pause, quietly, English): "I know. I saw the boxes." Sound: distant traffic, wind, a siren far away.
Why it works: Tone descriptors in the speaker labels steer the delivery, not just the words.
cinematic
21. Mountain rescue dog
A German shepherd in an orange rescue vest bounds through deep snow on a mountainside toward camera, a handler in red following behind. Camera: low-angle tracking shot moving backward in front of the dog, steady. Lighting: bright overcast, snow glare, cold blues and orange accents. Mood: urgent, heroic. Sound: crunching snow, panting, wind, a distant shout.
Why it works: A backward tracking shot on a moving subject shows off Kling's motion quality with a single, stable move.
cinematic
22. Neon diner
A waitress in a pale pink uniform refills a coffee cup for a lone trucker at a neon-lit roadside diner at 3 a.m. Camera: slow dolly along the counter from the coffee pot to the trucker's face, off-center composition. Lighting: pink and blue neon from the window, warm light over the counter. Mood: lonely, cinematic, 1980s Americana. Sound: coffee pouring, fridge hum, rain outside.
Why it works: One dolly move linking two characters, with a named era, gives a strong mood shot in a single generation.
animation
23. Plush toy adventure
3D animated style, soft and tactile like a stop-motion film. A small knitted bear with a button nose climbs over a pile of picture books on a child's bed, reaches the top and waves. Camera: medium shot, slow tilt up following the climb. Lighting: warm bedside lamp, deep blue night outside the window. Mood: sweet, playful. Sound: soft thumps, a tiny triumphant squeak.
Why it works: Naming the texture and animation style keeps Kling from drifting toward photorealism.
animation
24. Anime cafe opening
2D anime style. A girl with short blue hair flips the sign on a small corner café door from "CLOSED" to "OPEN" and smiles at the street. Cherry blossom petals drift past. Camera: medium shot, static, then a gentle push-in on her smile. Lighting: soft spring morning, pastel pinks and blues. Mood: hopeful. Sound: a door bell chime, birds. Girl (bright, Japanese): "いらっしゃいませ！"
Why it works: A named 2D style, quoted sign text and a Japanese line use Kling's style control, text rendering and multilingual audio together.
From prompt to posted video
Paste the prompt, get the video, post it
Hooked Pro is $29 a month for 1,000 credits across Veo 3.1, Kling 3.0, Seedance 2.0, Nano Banana Pro and GPT Image 2, plus AI UGC ads, captions and TikTok slideshows in the same plan. Or pay $89 once for the Studio and run it on your own API keys, at the provider's price.
Better results
Tips that make a difference
Do
- Write the prompt as scene direction, not a list of objects: say who moves, how, and what the camera does while they move.
- Pick one pacing word per shot: 'slow' for calm or suspense, 'steady' for readable action, 'quick' for energy (Kling's guide uses these three).
- Use shot sizes on purpose: extreme close-up for details, medium close-up for dialogue, full body for clothing and action, establishing wide for location.
- For multiple speakers, label each line: 'Mia (warm, amused, English): "..."'. Clean speaker labels are how Kling assigns voices in native audio.
- Use Multi-Shot (Auto) when you only need coverage of one scene; switch to Custom Multi-Shot when each shot needs its own framing, move and duration.
- Bind a product or character as an Element (2-4 reference images) before writing the prompt, then refer to it by name in every shot.
- Add composition words such as 'centered', 'rule of thirds' or 'off-center' when placement matters, as in a product on a shelf.
- Put on-screen text in quotes and say where it sits ('the label reads "GLOW" in white serif letters') so the text rendering has something exact to render.
- Test in 720p without audio, then render the final in 1080p with native audio; the credit cost differs by resolution and audio.
Common mistakes
- Writing an image prompt (a static description) and expecting motion; Kling needs actions and camera behaviour
- Packing a whole story into one paragraph instead of splitting it into labelled shots
- Dialogue without speaker labels in a multi-character scene, which can put lines in the wrong mouth
- Stacking several camera moves in one shot (push-in plus orbit plus tilt), which makes framing unstable
- Describing a product differently in each shot instead of binding it as an Element
FAQ
Frequently asked questions
What is the latest Kling AI model?
Kling VIDEO 3.0 and VIDEO 3.0 Omni, launched on February 5, 2026. A faster variant, Kling 3.0 Turbo, followed in June 2026.
How long can a Kling 3.0 video be?
Between 3 and 15 seconds per generation. You can also split that time into several shots with Multi-Shot.
Does Kling 3.0 generate audio?
Yes. It generates dialogue, sound effects and ambience in the same pass, in Chinese, English, Japanese, Korean and Spanish. Label each speaker with tone and language so the right voice gets each line.
How do I keep a product or character the same across shots?
Bind it as an Element by uploading 2 to 4 reference images or a short video, then refer to it by name in every shot.
Can I use Kling on Hooked?
Yes. Kling 3.0 is one of the video models available on Hooked, so these prompts can be pasted in directly.
Keep exploring
Related prompt libraries and pricing
Sources
- https://kling.ai/blog/kling-ai-prompt-guide
- https://kling.ai/quickstart/klingai-video-3-model-user-guide
- https://kling.ai/quickstart/klingai-video-3-omni-model-user-guide
- https://ir.kuaishou.com/news-releases/news-release-details/kling-ai-launches-30-model-ushering-era-where-everyone-can-be
- https://www.globenewswire.com/news-release/2026/02/05/3232837/0/en/kling-ai-launches-3-0-model-ushering-in-an-era-where-everyone-can-be-a-director.html
- https://higgsfield.ai/creator-hub/changelog
Two ways in · pick yours
Every template here opens in the app — 1,600+ and counting.
Pay monthly and we run the AI for you, credits included — or pay $89 once, own the app for life and bring your own keys.
Pro
Most flexibleWe run the AI for you. No API keys, nothing to set up — open the app and generate.
1,000 credits a month · cancel anytime
Secure checkout by Stripe · Cancel anytime
- About 10 UGC ads or 250 images a month
- Every video format: UGC, avatars, slideshows…
- 83 AI avatars & 76 voices in 25 languages
- Publish to TikTok, Instagram & YouTube
- 3 channels, 3 auto-channels, API included
Need more? One-time credit packs from $19. They never expire.
4.9/5 · 27,000+ users create with Hooked