Nano Banana Works is our sister publication. References to its still-image workflows below are owned-site recommendations.
Use-case prompts · wedding highlights
AI Wedding Video Prompts for H3 Max, Hailuo and Seedance
Treat a wedding film as short scene beats, not one long shot: first look, vows, first dance, toast, exit, venue establish. Copy a preset, swap the placeholders, then paste into MiniMax H3 Max, Hailuo, or fal.
Do not upload photos or generate recognizable footage of real people without their permission — couple, guests, vendors, and children included. These prompts use placeholders such as [PARTNER A] so you can describe wardrobe and setting without locking a real face. If you start from a still (image-to-video), only use images you have the right to use.
How wedding beats map to short clips
H3 Max-class models are strongest on 5–8 second clips with one camera idea. A highlight reel is a sequence of those beats, not a 10-minute locked narrative in one prompt.
Replace bracketed placeholders before you generate. Keep duration short while testing (5–8s @ 768P), then lengthen winners.
Longer films vs beat prompts
Longer, reference-locked wedding films (same faces and wardrobe across many shots) often need a different class of tool — for example Seedance-class models that hold identity from stills over a longer cut. This page still helps you write the beat briefs you paste into H3 Max, Hailuo, or fal for draft clips, social teasers, and mood tests. For CCTV-orbit, bullet-time, and product-drop briefs, see Seedance 2.5 bullet-time video prompts. If your old Sora wedding paragraph tried to be a whole film, remap the camera language on Sora → Seedance / Kling / Veo prompt remap (API hard-stop 24 Sep 2026).
Copy-ready wedding prompts
Each block is one H3 Max-style brief. Copy it as-is, or send the same beat into the prompt builder to tweak camera and duration.
Two partners in wedding attire at opposite ends of a quiet [VENUE] hallway with tall windows; [PARTNER A] in [ATTIRE A], [PARTNER B] in [ATTIRE B], both facing away until the turn. They slowly turn for a first look; [PARTNER A] covers their mouth, [PARTNER B] steps forward; dust motes drift in window light. Slow dolly forward. Soft golden hour, shallow depth of field. Aspect ratio 16:9, duration 8 seconds, resolution 768P. Sound: distant birds, fabric rustle, a soft inhale, no music.
A small outdoor ceremony at [VENUE]: [PARTNER A] and [PARTNER B] face each other at the altar, an officiant softly out of focus, seated guests blurred behind them. [PARTNER A] speaks vows while [PARTNER B] listens, eyes shining; a breeze moves [GARLAND / VEIL]; no guest faces in sharp focus. Static locked-off wide. Soft golden hour, shallow depth of field. Aspect ratio 16:9, duration 8 seconds, resolution 768P. Sound: quiet vows, a sniffle, leaves in the breeze, no score.
[PARTNER A] and [PARTNER B] slow-dancing in a dim [VENUE] reception room under warm string lights, other guests kept as soft bokeh in the background. They turn together in a close hold; [PARTNER A] laughs once; the camera completes a gentle orbit as the dress / suit hem moves. Slow orbit around subjects. 35mm anamorphic, practical lights only. Aspect ratio 16:9, duration 8 seconds, resolution 768P. Sound: shoe scuffs, low room murmur, faint unlicensed instrumental only, no recognizable commercial song.
A reception table at [VENUE]; a speaker in [OUTFIT] raises a glass, the couple seated nearby in soft focus, fairy lights and candles in the background. The speaker lifts a champagne flute, guests lean in, glasses catch highlights; keep faces generic unless you have consent to depict real people. Handheld phone-camera look. 35mm anamorphic, practical lights only. Aspect ratio 16:9, duration 6 seconds, resolution 768P. Sound: glasses clink, muffled speech, warm laughter, no music.
[PARTNER A] and [PARTNER B] leaving [VENUE] under a clear umbrella in light rain, wet pavement reflecting warm doorway light, guests as silhouettes tossing petals. They duck through the rain, laugh, and hurry toward a waiting car; petals stick to the umbrella; headlights bloom in the drizzle. Slow dolly forward. Moody cyan-orange cinematic grade. Aspect ratio 16:9, duration 8 seconds, resolution 768P. Sound: rain on umbrella, cheers, wet footsteps, a car door, no music.
Aerial establish of [VENUE] at golden hour: garden paths, stone facade, string lights just warming up, empty ceremony chairs in neat rows. A slow descending orbit reveals the grounds; no readable faces; a single figure in the distance is optional and tiny. Orbiting drone shot. Soft golden hour, shallow depth of field. Aspect ratio 16:9, duration 8 seconds, resolution 768P. Sound: wind across trees, distant birds, faint traffic, no music.
Beat 7 · detail insert
Ring and boutonniere macro
Copied
Placeholders: [RING STYLE], [FLOWER], [FABRIC]
Extreme close-up of a [RING STYLE] wedding band resting on a [FLOWER] boutonniere pinned to [FABRIC], window light raking across metal and petals. The camera pushes slowly into the engraving / stone as a single water bead slides off a leaf. Macro push into detail. Soft golden hour, shallow depth of field. Aspect ratio 16:9, duration 5 seconds, resolution 768P. Sound: almost silence, faint cloth rustle, no music.
Beat 8 · getting ready
Veil and window light
Copied
Placeholders: [PARTNER A], [ATTIRE], [ROOM]
[PARTNER A] in [ATTIRE] standing beside a tall window in a quiet [ROOM], a veil or jacket held just out of frame by an unseen helper. Hands smooth the fabric; lace or lapel catches sidelight; face stays three-quarter and softly lit, not a portrait lock. Slow dolly forward. Soft golden hour, shallow depth of field. Aspect ratio 16:9, duration 6 seconds, resolution 768P. Sound: distant hallway murmur, fabric, no music.
Beat 9 · sparkler send-off
Night sparkler corridor
Copied
Placeholders: [PARTNER A], [PARTNER B], [VENUE]
Night send-off at [VENUE]: [PARTNER A] and [PARTNER B] run down a corridor of sparklers held by silhouetted guests, embers drifting, warm practicals only. Handheld follow as sparks fall past the lens; faces of guests stay unreadable. Handheld phone-camera look. Moody cyan-orange cinematic grade. Aspect ratio 16:9, duration 6 seconds, resolution 768P. Sound: crackle of sparklers, cheers, footsteps on gravel, no music.
Beat 10 · social teaser
9:16 first-dance teaser
Copied
Placeholders: [PARTNER A], [PARTNER B], [VENUE]
Vertical frame: [PARTNER A] and [PARTNER B] in a close first-dance hold at [VENUE], string lights above, floor reflecting warm bulbs. Slow rise from waist-level to faces as they turn once; guests are bokeh only. Slow dolly forward. 35mm anamorphic, practical lights only. Aspect ratio 9:16, duration 5 seconds, resolution 768P. Sound: low room tone, shoe scuffs, no recognizable commercial song.
Paste into the H3 Max builder
The home text-to-video prompt generator uses the same fields (subject, action, camera, look, duration, sound). Wedding chips on that page fill first look, vows, first dance, toast, rainy exit, and drone venue — then you can generate and copy.
Only with consent from everyone whose face is recognizable, including guests who might appear in the frame. Prefer placeholders and original stills you own, or hire talent who agreed to AI use. Children and bystanders should not be uploaded casually.
Why short clips instead of one wedding-film prompt?
H3 Max-class text-to-video is built for short shots with one camera move. A highlight is a playlist of beats (venue, vows, dance, exit). Stitch those clips in an editor rather than asking one prompt for a full ceremony.
When would I use a Seedance-class model instead?
When you need longer, reference-locked footage that keeps the same faces and wardrobe across many shots. This page still helps you write the beat prompts you test first on H3 Max, Hailuo, or fal. Orbit and product-drop language lives on AI bullet-time video prompts.
Where do I run these prompts?
Common options include the fal H3 Max playground/API and MiniMax / Hailuo surfaces. This site does not generate video — it gives you copy-ready text to paste there.
Should the prompt name a first-dance song?
Avoid naming commercial recordings. Describe diegetic room sound, or a generic unlicensed instrumental, so the brief stays factual and you are not implying a licensed track.