PROMPT LIBRARY

Video prompts for the models leading the arena.

Each prompt is quoted word for word from the model maker's guide or from a creator's post, next to the output it produced. Click a picture to watch or view the original.

Maintained on GitHub: awesome-image-video-prompts awesome-prompts

Ultra close fashion portrait with candid flashes

Ultra close fashion portrait with candid flashes

Wan 3.0 · Text to video · 1,439likes · EN

The prompt splits the 15 seconds into five timed beats, each with its own camera move, gaze and flash cue. It also lists real skin flaws and bans plastic skin, which keeps the face photographic.

Source: @leaf_sanren on X
Night out friends and a quay leap

Night out friends and a quay leap

Wan 3.0 · Text to video · 1,353likes · EN

Opens with a style and characters block and a setting block, then a timestamped shot list that gives each beat its own camera move such as a backward track, a whip tilt and a high angle look down. A separate audio block asks for live sync sound with no music, which supports the raw phone footage look.

Source: @mightyking on X
Re-shoot your clip from new camera angles, same action

Re-shoot your clip from new camera angles, same action

Wan 3.0 · Video edit · 1,170likes · EN

Upload one continuous shot as the action reference and Wan re-films it from new angles. Regenerate if a shot drifts.

Source: @Framer_X on X
Man in camel coat at cafe counter

Man in camel coat at cafe counter

Wan 3.0 · Reference to video · 878likes · EN

A three panel character sheet is built first from one reference image, with front and back views on the left and one detailed portrait on the right. The front view face is blurred on purpose so the portrait is the only face reference, and outfit, fabric and film look are spelled out to keep the character consistent across scenes.

Source: @ivanka_humeniuk on X
Hand-drawn anime fight against a giant monster

Hand-drawn anime fight against a giant monster

Wan 3.0 · Reference to video · 502likes · EN

Two tagged references (hero, monster) and an instruction to keep their hand-drawn line work. Wan has its own anime look.

Source: @aimikoda on X
Four armed men, a silent standoff

Four armed men, a silent standoff

Wan 3.0 · Text to video · 261likes · EN

Written in numbered beats with restrained performances. It ends right before anyone fires.

Source: @VictorInFocus on X
Jewelled creatures turn into high heels

Jewelled creatures turn into high heels

Wan 3.0 · Text to video · 186likes · EN

A repeating transformation: each step onto a creature turns it into a shoe. The pattern is stated once and the model repeats it.

Source: @TaliaAariz on X
Shadow-puppet theatre battle with a pair of scissors

Shadow-puppet theatre battle with a pair of scissors

Wan 3.0 · Text to video · 184likes · EN

Starts with a style block (CG, lighting, rendering terms), then the fight, then two plain bans: no music, no subtitles.

Source: @FellMentKE on X
Steampunk detective montage across five locations

Steampunk detective montage across five locations

Wan 3.0 · Text to video · 152likes · EN

Opens mid-motion and keeps one sound element (a low cello) under every cut to hold the montage together.

Source: @socialwithaayan on X
Time-travelling train, one 30-second take

Time-travelling train, one 30-second take

Wan 3.0 · Text to video · 129likes · EN

Opens with a spec sheet (duration, ratio, format, style), then lists physical rules the scene must obey, such as passengers never teleporting.

Source: @johnAGI168 on X
Kung fu duel in a temple courtyard at dusk

Kung fu duel in a temple courtyard at dusk

Wan 3.0 · Text to video · 114likes · EN

Bracketed sections (`[PROJECT TYPE]`, `[Global Setting]`, …) and a closing checklist of what the fight must show.

Source: @LudovicCreator on X
Tokyo travel vlog in one-second beats

Tokyo travel vlog in one-second beats

Wan 3.0 · Text to video · 421likes · 中文

Chinese description with Japanese dialogue. Every second gets a location, an action, a line and the transition into the next one.

Source: @joshesye on X
Cold face on video, sweet smile in the photos

Cold face on video, sweet smile in the photos

Wan 3.0 · Reference to video · 240likes · 中文

Chinese prompt with labelled blocks. The joke is the hard cut between a cold moving shot and a smiling still photo, and the rules forbid any fade between them.

Source: @johnAGI168 on X
Finger dance in one locked-off take

Finger dance in one locked-off take

Wan 3.0 · Reference to video · 194likes · 中文

Names the character once (`粉宝`) and refers back to her by name. It even specifies which hand does which move.

Source: @johnAGI168 on X
A street chase where photo-editing tools are real objects

A street chase where photo-editing tools are real objects

Wan 3.0 · Text to video · 189likes · 中文

Cursors, selection boxes and transform handles exist on the street with shadows and occlusion. Long ban list keeps it from turning cartoony.

Source: @johnAGI168 on X
Twenty shot character detail montage, face revealed last

Twenty shot character detail montage, face revealed last

MiniMax H3 · Reference to video · 3,281likes · EN

A numbered shot list with time blocks and one abrupt camera move per shot builds three connected action bursts. Faces stay out of frame until shot 20, and hard match cuts link similar shapes without morphing.

Source: @aimikoda on X
Anime boy glancing back while running in park

Anime boy glancing back while running in park

MiniMax H3 · Reference to video · 1,599likes · EN

A 3x3 panel image is fed as the reference, and the prompt asks for cuts running from the first panel to the last, so the storyboard plays back almost as drawn. Forbidding multiple panels on screen and BGM keeps each shot full frame and the audio clean.

Source: @8co28 on X
Flat Vector Curry Delivery Explainer Animation

Flat Vector Curry Delivery Explainer Animation

MiniMax H3 · Image to video · 766likes · EN

The key visual is locked as the exact first frame, and five timed shots each name their background color, motion type and on screen text. Exact Japanese strings, a ban on extra text or UI, and a fixed end hold at 12 seconds keep the typography clean.

Source: @ponzponz15 on X
Make the man in an existing clip speak, in a borrowed voice

Make the man in an existing clip speak, in a borrowed voice

MiniMax H3 · Video edit · EN

Edit-mode prompt with six fields. `retention_analysis` says what to keep from each input: the source video fully, its music partly, and only the voice timbre from the second audio clip.

Source: MiniMax H3 model card
Ramen in front, family dinner behind, one focus pull

Ramen in front, family dinner behind, one focus pull

MiniMax H3 · Image to video · EN

The first line pins the input photo to frame 0, then the prompt describes the whole eight seconds as one locked-off shot whose only move is the focus shifting from the bowl to the family.

Source: MiniMax H3 model card
Starship bridge, the fleet jumps to hyperspace

Starship bridge, the fleet jumps to hyperspace

MiniMax H3 · Text to video · EN

The full H3-Context-IR format: numbered shots with a timed cut at 00:04.500, then separate fields for sound and score. Most local H3 runs skip this rewriting step, which is why they look flat.

Source: MiniMax H3 model card
Seoul subway revenge comedy

Seoul subway revenge comedy

Seedance 2.5 · Text to video · 10,056likes · EN

Sets the camera language in the first line (handheld, fluorescent, real time), then plays the gag in clear beats and spells out what must stay visible at the end.

Source: @AIwithSynthia on X
A summer evening in Seoul, shot like a home video

A summer evening in Seoul, shot like a home video

Seedance 2.5 · Text to video · 6,187likes · EN

Pure text, no reference image. Realism comes from asking for small accidents and imperfect camcorder operation.

Source: @AIwithkhan on X
Katsudon cooked in fast-cut anime style

Katsudon cooked in fast-cut anime style

Seedance 2.5 · Text to video · 4,935likes · EN

Opens with an explicit ban on imitating any existing anime, then walks through the recipe step by step so every cut has a job.

Source: @Goodmanprotocol on X
Rainy morning, coffee, walk to university

Rainy morning, coffee, walk to university

Seedance 2.5 · Text to video · 4,181likes · EN

Under 1,000 characters for 30 seconds. A tight routine plus a short list of failure modes to avoid.

Source: @nawalsehar on X
Cafe table talk reshot from thirteen new angles

Cafe table talk reshot from thirteen new angles

Seedance 2.5 · Video edit · 3,287likes · EN

The prompt locks performance, audio, lighting and tempo of the source video, then varies only camera position and lens through a timed shot list. Measured speech windows, an eyeline rule and an empty sky rule keep lip sync and background clean in every new angle.

Source: @Scenario_gg on X
1950s diner: time freezes, then rewinds

1950s diner: time freezes, then rewinds

Seedance 2.5 · Text to video · 2,873likes · EN

A single effect carried for 30 seconds. The diner is described once in detail so the freeze and rewind read clearly against it.

Source: @techhalla on X
Lab containment breach, sci-fi horror

Lab containment breach, sci-fi horror

Seedance 2.5 · Text to video · 2,425likes · EN

Asks for an immediate hook and grounded practical effects, which keeps the creature from looking like CGI.

Source: @auqibhabib on X
A woman stands still while the crowd moves in sync

A woman stands still while the crowd moves in sync

Seedance 2.5 · Reference to video · 2,128likes · EN

Uploaded images lock the woman, the crowd and the framing. The text only directs motion and lists what must not appear.

Source: @AIwithkhan on X
Friends film a night out on a phone

Friends film a night out on a phone

Seedance 2.5 · Text to video · 2,038likes · EN

Starts by naming the camera operator (a friend walking with the group). Everything about the shake and framing follows from that.

Source: @mightyking on X
Crying ginger cat aims handgun at zombie calico

Crying ginger cat aims handgun at zombie calico

Seedance 2.5 · Text to video · 2,000likes · EN

A timestamped shot list assigns each beat a framing, an action and an emotional turn, so the tension in the dark cabin resolves cleanly into the golden hour meadow. The opening lays out both characters, both settings and the lighting contrast between the two halves, which keeps the cats and the locations consistent across cuts.

Source: @ElsaSofia__AI on X
Anime character reaching past black letterbox bars

Anime character reaching past black letterbox bars

Seedance 2.5 · Reference to video · 1,675likes · EN

The prompt states a back to front compositing order (background, black bars, character) so the foreground character overlaps the bars while the bars stay fixed. A stationary wide angle camera, an opening lunge toward the lens and a closing instruction not to crop the character inside the window make the overlap repeat. Bar size still varies between generations.

Source: @aimikoda on X
A woman teaches a dragon to bake heart-shaped bread

A woman teaches a dragon to bake heart-shaped bread

Seedance 2.5 · Text to video · 1,430likes · EN

Short and specific: colours of the dragon, one hard rule (all bread heart-shaped), and a mood word to end on.

Source: @ElsaSofia__AI on X
Toddler copies dad's shaving with soap foam

Toddler copies dad's shaving with soap foam

Seedance 2.5 · Text to video · 1,379likes · EN

Vertical 3D family comedy. One clear setup, one gag, one laugh.

Source: @Zarnab_with_Ai on X
Celestial puppeteer rebuilds a collapsing golden star

Celestial puppeteer rebuilds a collapsing golden star

Seedance 2.5 · Reference to video · 1,127likes · EN

A fixed painterly style block with explicit negatives opens and closes the prompt, then WORLD and CHARACTER blocks lock the look. Ten timed shots of three seconds each, with the shot type named in every cut, give the 30 second piece a clear arc from macro detail to hero frame. The character sheet made with GPT Image 2.5 serves as the identity reference.

Source: @itsPixieVerse on X
Friday morning walk in an old Seoul neighbourhood

Friday morning walk in an old Seoul neighbourhood

Seedance 2.5 · Reference to video · 1,124likes · EN

Character comes from a reference image. The closing list of bans (beauty filter, changing outfit) keeps her stable for 30 seconds.

Source: @AIwithkhan on X
Woman and Pet Dragon Cook a Roast

Woman and Pet Dragon Cook a Roast

Seedance 2.5 · Text to video · 1,043likes · EN

A second-by-second shot list from wide shot to extreme close-up builds a full comedy arc in 30 seconds, from the dragon's fire blast to a gentle forgiveness beat. Character sheets for both leads, a setting list and a separate audio line for fire, steam and dragon sounds keep the look and sound consistent.

Source: @Alina_with_Ai on X
Selfie-style skincare UGC ad

Selfie-style skincare UGC ad

Seedance 2.5 · Reference to video · 1,017likes · EN

Third step of a thread: a character sheet and a product shot go in as references, and the prompt insists `Not an ad` to get the self-filmed feel.

Source: @abxxai on X
Zombie outbreak in a busy shopping mall

Zombie outbreak in a busy shopping mall

Seedance 2.5 · Text to video · 1,002likes · EN

Starts in normal trading hours so the outbreak has something to break. Ends on a quiet beat instead of an explosion.

Source: @auqibhabib on X
Early-2000s Korean birthday on a DV camcorder

Early-2000s Korean birthday on a DV camcorder

Seedance 2.5 · Text to video · 918likes · EN

Period realism by naming the device. Scenes are numbered and the prompt demands they play in that exact order.

Source: @saniaspeaks_ on X
Masked warrior fights scythe reaper in desert dust

Masked warrior fights scythe reaper in desert dust

Seedance 2.5 · Reference to video · 882likes · EN

Two image references lock the reaper and the warrior character sheet, then a timed shot list from 0 to 30 seconds names a camera move and a clear action beat for each segment. A shared style block of anamorphic lens, haze and particle physics keeps the look consistent across the cuts.

Source: @techhalla on X
Baby Dragon Sneezes Fire in Winter Cabin

Baby Dragon Sneezes Fire in Winter Cabin

Seedance 2.5 · Text to video · 833likes · EN

The prompt splits 25 seconds into four timed scenes, each with its own shot size and one clear action beat. A blue to amber color shift tracks the fire lighting up, and the sound design and a negative prompt keep the creature consistent.

Source: @Alina_with_Ai on X
Mini DV coffee ASMR vlog

Mini DV coffee ASMR vlog

Seedance 2.5 · Text to video · 833likes · EN

Organised under labelled blocks (`CAMERA / LOOK`, then subject, action, sound). Easy to reuse as a template.

Source: @Strength04_X on X
Painted Flame Karate Kata in Blue Void

Painted Flame Karate Kata in Blue Void

Seedance 2.5 · Reference to video · 802likes · EN

Two labeled references split identity from body, costume and effects, and a sixteen shot timeline gives each cut a time range, camera move and technique. Sound cues are tied to the cuts and an avoid list keeps the painted look free of glossy CG fire.

Source: @aimikoda on X
Hiding three dragon hatchlings from a hunter

Hiding three dragon hatchlings from a hunter

Seedance 2.5 · Text to video · 781likes · EN

Written as a sequel: the first line states it continues an earlier clip, which helps keep tone and characters consistent across a series.

Source: @Adnan_Ai2 on X
Airport immigration makeup wipe comedy in twelve shots

Airport immigration makeup wipe comedy in twelve shots

Seedance 2.5 · Text to video · 740likes · EN

A character block repeated in every part locks faces, outfits and location, and the 12 shots carry exact timestamps and panel labels. Each shot states framing and camera behavior, and a master effects list plus an energy arc keep the 30 second real time pacing.

Source: @ElsaSofia__AI on X
Outfit changes in one take, with Chinese dialogue

Outfit changes in one take, with Chinese dialogue

Seedance 2.5 · Reference to video · 676likes · EN

Image 1 locks only the face and hair, and the clothing is described per change. It also tells the lead to close her mouth when the off-screen man speaks.

Source: @cnyzgkc on X
Two-idol K-pop music video, shot by shot

Two-idol K-pop music video, shot by shot

Seedance 2.5 · Text to video · 547likes · EN

A 30-second MV laid out in 2 to 4 second time ranges, alternating between three sets.

Source: @Just_sharon7 on X
Woman fights three men in a restaurant kitchen

Woman fights three men in a restaurant kitchen

Seedance 2.5 · Text to video · 514likes · EN

Ten seconds of fight choreography with the grade and camera fixed up front and a sound list at the end, no music.

Source: @Alina_with_Ai on X
Two Anime Fighters Brawl on Rooftop

Two Anime Fighters Brawl on Rooftop

Seedance 2.5 · Reference to video · 1,962likes · 中文

Character reference blocks lock each fighter's outfit, then a style and energy header sets cel shading, impact frames and constant camera motion. A four shot list gives each shot its own camera path and a chain of strikes with no pauses, and an Avoid line bans slow motion and freeze frames.

Source: @aimikoda on X
Bird leaps from laptop photo onto hand

Bird leaps from laptop photo onto hand

Gemini Omni Flash · Video edit · 1,385likes · EN

The prompt first locks the screen content to match the source video, then describes one continuous action in order: head turn, leaving the photo, landing on the hand. A real filmed setup gives the effect a physical anchor.

Source: @alexanderchen on X
Glossy 3D words appearing one at a time

Glossy 3D words appearing one at a time

Gemini Omni Flash · Text to video · 1,219likes · EN

The exact word sequence is spelled out, with one word on screen at a time. Asking for a different animated style per word and pacing to a rhythm gives the model clear control over timing and text motion.

Source: @Google on X
Miniature farmhouse foundation built by giant hands

Miniature farmhouse foundation built by giant hands

Gemini Omni Flash · Text to video · 1,205likes · EN

The prompt lists concrete miniature materials and tools, then states the construction stage and the empty pool area, so the model knows what is built and what is not. Macro lens, shallow depth of field and giant hands against tiny objects set the scale contrast.

Source: @Strength04_X on X
Kitchen frying pan from a commercial storyboard

Kitchen frying pan from a commercial storyboard

Gemini Omni Flash · Image to video · 589likes · EN

A structured prompt written as labeled sections (format, header, panel layout, photography style, footer) makes GPT Image 2 produce an eight frame storyboard board. That board then serves as the visual reference that Gemini Omni Flash animates into a product shot.

Source: @Strength04_X on X
A flock of birds forms the shape of a bird, to music

A flock of birds forms the shape of a bird, to music

Gemini Omni Flash · Reference to video · EN

Three inputs, each with its own tag: `<video>` for motion, `<image>` for shape, `<audio>` for timing.

Source: Gemini Omni prompt guide
Add cartoon motion lines to live skateboarding

Add cartoon motion lines to live skateboarding

Gemini Omni Flash · Video edit · EN

`keeping everything the same` locks the footage, so the edit only layers effects on top.

Source: Gemini Omni prompt guide
Apartment lights switch on to the beat

Apartment lights switch on to the beat

Gemini Omni Flash · Video edit · EN

Syncs a visual event to an audio track by naming both in one sentence. No timestamps needed.

Source: Gemini Omni prompt guide
Butterfly to bee to fireflies, one change per turn

Butterfly to bee to fireflies, one change per turn

Gemini Omni Flash · Video edit · EN

Two short edit turns on the same clip. Each turn names one subject and its replacement, and everything else stays put.

Source: Gemini Omni prompt guide
Hand it a storyboard, get the scene

Hand it a storyboard, get the scene

Gemini Omni Flash · Reference to video · EN

The storyboard image carries the plot. The text only sets reading order, total length and tone.

Source: Gemini Omni prompt guide
Harbour conversation extended twice

Harbour conversation extended twice

Gemini Omni Flash · Extend · EN

Each extension only says what happens next: one line of dialogue, then a camera pull-back and a music cue.

Source: Gemini Omni prompt guide
Kinetic type, one word per beat

Kinetic type, one word per beat

Gemini Omni Flash · Text to video · EN

Lists the exact words in order and asks for a new animation style on each. Good test of on-screen text.

Source: Gemini Omni prompt guide
Move the camera over the violinist's shoulder

Move the camera over the violinist's shoulder

Gemini Omni Flash · Video edit · EN

A camera change written as one plain sentence. Omni re-renders the same performance from the new angle.

Source: Gemini Omni prompt guide
One clip through crayon, pencil, glass and risograph

One clip through crayon, pencil, glass and risograph

Gemini Omni Flash · Video edit · EN

Restyles a reference video in four stages. Each stage gets its own texture words, so the transitions read clearly.

Source: Gemini Omni prompt guide
Quantum computing explained in risograph flat style

Quantum computing explained in risograph flat style

Gemini Omni Flash · Text to video · EN

Asks for an explanation and leaves the content to the model's world knowledge. The prompt only pins down the visual style.

Source: Gemini Omni prompt guide
Ships to origami, astronaut to anemone, ships to stingrays

Ships to origami, astronaut to anemone, ships to stingrays

Gemini Omni Flash · Video edit · EN

Three edits in a row on one shot. The scene and camera stay consistent across every turn.

Source: Gemini Omni prompt guide
Start on the shoes, tilt up, then widen

Start on the shoes, tilt up, then widen

Gemini Omni Flash · Video edit · EN

Three camera moves chained in one line, in the order they happen.

Source: Gemini Omni prompt guide
Whip-pan from drummer to saxophonist and dancer

Whip-pan from drummer to saxophonist and dancer

Gemini Omni Flash · First and last frame · EN

First and last frames fix both ends. The prompt describes the move between them and closes with `One continuous shot, no jump cuts`.

Source: Gemini Omni prompt guide
Macro shot of a vintage wristwatch between fingers

Macro shot of a vintage wristwatch between fingers

Kling 3.0 · Text to video · 2,534likes · EN

The prompt fixes the camera distance in inches and lists concrete surface details such as dust, micro scratches and patina, so the model holds texture. A single warm grazing key light, a shallow depth of field and one small motion, the second hand, keep the shot stable.

Source: @higgsfield_ai on X
Basketball player backflip dunk on outdoor court

Basketball player backflip dunk on outdoor court

Kling 3.0 · Text to video · 2,533likes · EN

The prompt splits one action into four numbered shots (dribble, jump, mid-air flip, dunk), each with its own shot size and angle. Low angles and a slow-motion close-up on the flip give Kling's physics a clear beat to simulate.

Source: @higgsfield_ai on X
Woman slices wooden dummy in night courtyard

Woman slices wooden dummy in night courtyard

Kling 3.0 · Text to video · 2,473likes · EN

Six numbered shots, each with its own framing, move a single scene from a macro mouth detail through wide action to a closing portrait. Naming the camera angle and the physical effect per shot, such as flying wood splinters and dust, gives the model clear impact beats.

Source: @higgsfield_ai on X
Golden eagle POV flight over mountain valleys

Golden eagle POV flight over mountain valleys

Kling 3.0 · Text to video · 2,454likes · EN

The camera is locked to the eagle's head, with wing and feather edges kept in frame as a stable anchor. Ground parallax cues (rivers, valleys, peaks) and the constraints on jitter, motion blur and cuts keep the flight physically coherent in one continuous shot.

Source: @higgsfield_ai on X
Turntable Push-In to Stylus Needle Macro

Turntable Push-In to Stylus Needle Macro

Kling 3.0 · Image to video · 2,357likes · EN

Starts from a given frame and runs one continuous take. A rigid bolt-mounted push-in locks onto the needle tip, and the prompt lists macro details, dominant sound design and explicit camera rules.

Source: @higgsfield_ai on X
Panicked woman in striped robe on Manhattan street

Panicked woman in striped robe on Manhattan street

Kling 3.0 · Text to video · 622likes · EN

Each shot gets its own number, with a stated shot size, lens behavior and one concrete action. Cutaways to macro details and a contrasting calm scene build rhythm, and the character's look is repeated so she stays consistent across shots.

Source: @higgsfield_ai on X
Low angle tracking shot locked on paper bag

Low angle tracking shot locked on paper bag

Kling 3.0 · Text to video · 554likes · EN

The prompt names one anchor object and states that the camera is rigidly locked to it, so the subject stays centered while the background streams past. Concrete motion and texture cues such as pumping arm, whitening knuckles and vibrating paper tie the subject's movement to the shot.

Source: @higgsfield_ai on X
Billiard ball POV rolling into corner pocket

Billiard ball POV rolling into corner pocket

Kling 3.0 · Text to video · 449likes · EN

The prompt names an unusual camera identity, the ball itself, and splits the shot into strike, roll and drop beats. Low table level framing, motion blur, lamp light streaks and parallax from rails and background players keep the POV motion physical.

Source: @higgsfield_ai on X
Hotel corridor fight under red neon lights

Hotel corridor fight under red neon lights

Kling 3.0 · Text to video · 445likes · EN

The prompt sets the location and lighting first, then lists the action beats in order: stance, entrance from the right, block, grab, knee strike, push to the wall. A closing camera line asks for a steady tracking move that follows the action.

Source: @higgsfield_ai on X
Woman dancing in neon nightclub and rain

Woman dancing in neon nightclub and rain

Kling 3.0 · Text to video · 445likes · EN

The prompt stacks subject, wardrobe, setting, lighting palette and camera moves in one dense paragraph. A separate negative list and a closing technical line for resolution, frame rate and grade keep the look consistent.

Source: @CaliraVal on X
Asking directions in Madrid, in halting Spanish

Asking directions in Madrid, in halting Spanish

Kling 3.0 · Image to video · EN

Three speakers with different fluency: the tourist is told to speak slowly with an awkward accent, the clerk lightly and cheerfully.

Source: Kling VIDEO 3.0 user guide
Bento handed over at a misty station, in Japanese

Bento handed over at a misty station, in Japanese

Kling 3.0 · Image to video · EN

Japanese lines in corner brackets, no speaker labels beyond `the boy` and `the girl`. Shows Kling's Japanese lip sync.

Source: Kling VIDEO 3.0 user guide
Camera circles to her face, she looks up and smiles

Camera circles to her face, she looks up and smiles

Kling 3.0 · Image to video · EN

Same one-line prompt run twice, with and without a bound element. The page shows both outputs so you can compare face consistency.

Source: Kling VIDEO 3.0 user guide
Four family members react to a plot twist

Four family members react to a plot twist

Kling 3.0 · Image to video · EN

Four speakers, each tagged by role and tone (`Mom (softly, in a surprised tone)`), so lines land on the right face.

Source: Kling VIDEO 3.0 user guide
Inside a plaster-statue tower, a dinosaur reunion

Inside a plaster-statue tower, a dinosaur reunion

Kling 3.0 · Image to video · EN

Declares `single unbroken shot with no edited transitions` up front, then carries one emotional arc from despair to reunion.

Source: Kling VIDEO 3.0 user guide
Joseon noblewoman and her maid

Joseon noblewoman and her maid

Kling 3.0 · Image to video · EN

Period drama in Korean. Tone words (`warm, gentle`, `with deference`) set the social register between the two.

Source: Kling VIDEO 3.0 user guide
Korean high-school rooftop at night

Korean high-school rooftop at night

Kling 3.0 · Image to video · EN

Language is declared inside each speaker tag: `Boy (casual tone, Korean)`.

Source: Kling VIDEO 3.0 user guide
One take follows a woman from elevator to desk

One take follows a woman from elevator to desk

Kling 3.0 · Image to video · EN

Start frame plus a bound character element. The prompt spells out how the camera moves with her: track when she walks, stop when she stops.

Source: Kling VIDEO 3.0 user guide
Ordering dinner with an Indian English accent

Ordering dinner with an Indian English accent

Kling 3.0 · Image to video · EN

Accent is named once before the line. The pause between the two sentences is written as an action (`then he looks up`).

Source: Kling VIDEO 3.0 user guide
Perfume ad by a Paris window

Perfume ad by a Paris window

Kling 3.0 · Image to video · EN

Product shot that keeps the bottle's printed name legible. Voiceover, accent and BGM are all specified in the prompt.

Source: Kling VIDEO 3.0 user guide
Snowmobile run in six shots

Snowmobile run in six shots

Kling 3.0 · Image to video · EN

Six shots, each defined only by angle and camera move. The landscape is described once, in the last shot.

Source: Kling VIDEO 3.0 user guide
Truck driver in four handheld shots

Truck driver in four handheld shots

Kling 3.0 · Image to video · EN

Custom Multi-Shot from Kling creative partner Dave Clark. Every shot repeats `cinematic handheld` so the look holds across cuts.

Source: Kling VIDEO 3.0 user guide
Villa terrace, a two-person summer dialogue

Villa terrace, a two-person summer dialogue

Kling 3.0 · Image to video · EN

Multi-Shot on: the model plans the shot-reverse-shot by itself from one paragraph of action and dialogue.

Source: Kling VIDEO 3.0 user guide
Woman in green runs across a moonlit garden, 15 s

Woman in green runs across a moonlit garden, 15 s

Kling 3.0 · Image to video · EN

A 15-second continuous shot written as a director's paragraph: lens, rig height, grade, then action in order.

Source: Kling VIDEO 3.0 user guide
Giant girl kneeling over a Tokyo crossing

Giant girl kneeling over a Tokyo crossing

Veo 3.1 · Image to video · 3,040likes · EN

A bracketed universal template fixes the camera angle, the giant's outfit, location and action, then adds tiny cars and people near the hands and feet to establish scale. Generate the still with Nano Banana Pro at 4K, then animate it with Veo 3.1 as image to video.

Source: @MrDavids1 on X
Empty room timelapse building a loft

Empty room timelapse building a loft

Veo 3.1 · First and last frame · 1,986likes · EN

Start and end frames, both made in Nano Banana Pro, fix the empty room and the finished design. The prompt only describes the fast motion construction between them, so Veo fills in the transformation.

Source: @venturetwins on X
Golfer on stormy cliff, camera tracks ball

Golfer on stormy cliff, camera tracks ball

Veo 3.1 · Text to video · 666likes · EN

The prompt runs as a beat by beat sequence: swing, camera locks onto the ball, approach, then an extreme close up at the cup. It states the tracking behavior outright, with the ball sharp and centered while the background streaks, and adds sound cues like the crack of the club and howling wind.

Source: @Dheepanratnam on X
Strawberry skincare jar assembling on red background

Strawberry skincare jar assembling on red background

Veo 3.1 · Reference to video · 582likes · EN

The prompt starts from an empty set and numbers each object's entrance, so the assembly order is explicit. A locked camera, a ban on bouncing and an end frame matching the reference image keep the motion clean and controlled.

Source: @ChillaiKalan__ on X
Excited Streamer Reacting at Neon Gaming Desk

Excited Streamer Reacting at Neon Gaming Desk

Veo 3.1 · Image to video · 366likes · EN

The prompt describes a medium close-up with pulsing neon light and a handheld micro-sway, so the shot feels like a live stream. It also lists audio layers such as room ambience, keyboard clicks, mic hum and distant game sounds. A Midjourney character reference and a Nano Banana Pro start frame keep the face consistent.

Source: @HBCoop_ on X
Crystal flowers on a moon-dust plain

Crystal flowers on a moon-dust plain

Veo 3 · Text to video · EN

Short world-building prompt. One striking detail (`the only footprints in untouched dust`) does more than a long list.

Source: Veo 3 prompt guide
Keyboard made of candy

Keyboard made of candy

Veo 3 · Text to video · EN

A separate `Audio:` line tells the model what the scene should sound like.

Source: Veo 3 prompt guide
Off-road rally through a river crossing

Off-road rally through a river crossing

Veo 3 · Text to video · EN

A play-by-play for an 8-second action beat. It also asks for unbranded vehicles, which keeps logos out of the frame.

Source: Veo 3 prompt guide
Old sailor speaks in verse

Old sailor speaks in verse

Veo 3 · Text to video · EN

Character first (hat, beard, pipe), then the spoken line in quotes. Veo 3 lip-syncs the quoted text.

Source: Veo 3 prompt guide
Stop-motion camper and a bear with a comeback

Stop-motion camper and a bear with a comeback

Veo 3 · Text to video · EN

The whole prompt is a format label plus two lines of dialogue. Veo fills in the scene.

Source: Veo 3 prompt guide
Anime woman wakes from cryosleep into forest

Anime woman wakes from cryosleep into forest

Sora 2 · Text to video · 2,957likes · EN

A style line up front sets the anime look, then timecoded Cut to beats every two seconds pin down each shot. Each beat names one action and one camera position, so the story moves from lab to ladder to forest.

Source: @EccentrismArt on X
A chubby purple monster comes out of the fridge

A chubby purple monster comes out of the fridge

Sora 2 · Image to video · EN

Two short beats: the door opens, the monster comes out.

Source: Sora 2 prompting guide
Edit the monster: recolour it, then add a second one

Edit the monster: recolour it, then add a second one

Sora 2 · Video edit · EN

Each edit is a separate run on the original clip and changes one thing. OpenAI's advice: describe only the tweak.

Source: Sora 2 prompting guide
She turns, smiles and walks out of frame

She turns, smiles and walks out of frame

Sora 2 · Image to video · EN

With an input image the first frame is set, so the prompt only describes what happens next.

Source: Sora 2 prompting guide

Prefer to learn the craft step by step? The 16 chapter course is here.