AI Video Prompts for Fashion Lookbook Videos

A fashion lookbook lives or dies on movement. A still image shows the cut; a lookbook video shows how the cut behaves — how the fabric falls when someone turns, how a hem catches the light, how a coat moves on a body walking toward you. That motion is exactly why video commerce converts: fashion brands using video see roughly 2.5x higher conversion than static product pages, because the viewer can finally judge fit and drape the way they would in a fitting room. AI video prompts are how you produce that footage at the speed a seasonal drop demands, without booking a studio, a model, and a steadicam operator for every SKU.

The catch is that fashion is unforgiving about two things: fabric and faces. A vague prompt gives you uncanny hands and fabric that defies physics. A precise prompt gives you something that looks like it was pulled from a brand’s spring campaign. Here’s how to write AI video prompts for fashion lookbook videos that respect the garment.

Why Fashion Prompts Are About Fabric and Motion

If you’ve written prompts for product video, fashion adds two demanding variables: the way material moves and the believability of the human wearing it. Telling a model “a woman in a dress” gives you almost nothing to work with. Telling it “a flowing silk slip dress that ripples as she turns” gives it the physics cues it needs to render drape convincingly.

Three things separate a usable fashion clip from an uncanny one:

• Fabric behavior — Name the material and how it moves: “structured wool coat that swings,” “sheer chiffon that floats,” “stiff denim that holds its shape.” Motion language is what teaches the model the physics.
• Garment-led framing — Frame for the clothing, not the face. Closer crops on detail, fuller shots for silhouette. This sidesteps the model’s weakest area (faces and hands) while showing what shoppers actually buy.
• Editorial mood — Lookbooks have a point of view. “Soft overcast daylight, muted earth tones” reads completely differently from “high-contrast studio strobe, glossy editorial.” Name the aesthetic.

The Fashion Lookbook Prompt Formula

Here’s the structure that holds up across Sora, Runway, Kling, Veo, and Luma:

[Garment: type, material, color, fit] [Motion: how the fabric behaves]
[Subject framing: full body / detail / silhouette] [Camera: lens, movement]
[Lighting: quality and direction] [Mood: editorial style, color palette]
[Aspect ratio] [Duration]

This extends the core prompt structure by putting the garment and its motion first — in fashion, the clothing is the subject, and everything else frames it.

Lookbook Prompt Templates You Can Use

1. The Walk-Toward (Silhouette Hero)

The signature lookbook shot. A model walks toward camera so the garment moves naturally with the body.

A model walks slowly toward the camera down a wide concrete
corridor, wearing an oversized camel wool coat that swings gently
with each step. The coat is unbuttoned over a cream knit. Full
body framing, low angle. Camera tracks backward to keep pace at
walking speed. Shot on 50mm lens, f/2.8. Soft directional
daylight from a tall window on the left. Minimal editorial mood,
muted neutral palette. 9:16. 6 seconds.

Swap the garment and palette for your collection. Keep “tracks backward at walking speed” — it’s what makes the fabric move believably.

2. The Turn (Drape Reveal)

A slow 180-degree turn shows how the garment reads from every angle. This is the clip that replaces a fitting room.

A model stands in a softly lit studio wearing a floor-length
emerald silk slip dress that ripples and catches the light as she
turns slowly in place. The fabric flows around her ankles. Medium
full shot. Camera holds static, slight push-in. Shot on 85mm
lens, f/2.0, shallow depth of field. Warm key light from the
right with a soft fill. Luxury editorial mood, rich saturated
color. 9:16. 6 seconds.

3. The Fabric Macro (Detail Beat)

Close enough to sell texture. Knitwear, leather, embroidery, hardware — the details that justify the price.

Extreme close-up tracking slowly across the shoulder seam of a charcoal herringbone wool blazer, revealing the weave texture, a horn button, and crisp lapel stitching. The fabric shifts subtly as the body moves. Shot on 100mm macro lens, f/2.8, razor-thin depth of field. Soft directional side light creating texture shadows. Tactile, premium editorial. 16:9. 5 seconds.

4. The Editorial Mood Frame (Lifestyle Context)

Places the garment in a world. This is the clip that sells the feeling of the collection, not just the item.

A model in a flowing ivory linen dress stands on a sun-bleached
stone balcony overlooking the sea, the breeze lifting the hem and
sleeves. Wide shot, model slightly off-center. Camera slowly
cranes up. Shot on 35mm lens, f/4.0. Bright Mediterranean midday
light, warm and airy. Relaxed resort-wear editorial, soft pastel
palette. 16:9. 7 seconds.

5. The Detail-in-Motion (Accessories)

Bags, shoes, jewelry — accessories need motion to show scale and how they sit on the body.

Close-up of a model's hand adjusting a quilted leather shoulder
bag as she walks, the gold chain strap catching the light and
swinging gently. The bag rests against a tailored trench coat.
Shot on 50mm lens, f/2.0, shallow depth of field. Soft overcast
daylight. Clean, minimal accessory editorial, neutral tones.
9:16. 5 seconds.

6. The Collection Sweep (Multi-Look)

A single moving shot past several looks — useful for a campaign opener or a “shop the collection” beat.

Slow lateral dolly gliding past three models standing against a
seamless beige backdrop, each wearing a different look from a
knitwear collection in coordinating earth tones. Camera moves
left to right at chest height. Full body framing. Shot on 35mm
lens, f/4.0. Even soft studio lighting. Cohesive editorial
campaign mood, warm neutral palette. 16:9. 7 seconds.

If hand-writing a prompt for every look in a 40-piece drop sounds like its own full-time job, LzyPrompt generates fashion-ready prompts tuned per model — describe the garment, its fabric, and your collection’s mood, and get a clip-ready prompt for Sora, Runway, Kling, or Veo. Generate your first prompt free.

Matching Lookbook Prompts to Generators

Each tool has a fashion personality worth knowing.

Kling is notably strong on human motion and fabric physics — walks, turns, and flowing material tend to hold together well. A first pick for the walk-toward and turn beats.

Runway Gen-4 keeps a garment and model consistent across clips, which matters when you’re building a multi-shot lookbook around the same look. Reach for it when continuity is the priority.

Sora handles editorial camera work — the crane, the lateral dolly, lighting that shifts as someone moves — making it strong for mood and lifestyle frames.

Veo 3 delivers photographic realism and natural skin and light, useful when a face is unavoidably in frame and needs to read as real.

Luma Dream Machine is a fast, reliable option for detail-in-motion clips where you mainly need smooth, clean movement.

Tips That Make Fashion Footage Believable

Always describe how the fabric moves. “Silk that ripples,” “wool that swings,” “denim that holds its shape.” Material plus motion is the single biggest upgrade to a fashion prompt — it gives the model the physics it can’t infer from a noun.

Frame for the garment, not the face. AI models are still weakest on faces and hands. Closer crops on fabric and detail, fuller shots for silhouette, and three-quarter angles all hide the model’s limitations while showcasing exactly what shoppers want to evaluate.

Commit to one lighting mood per collection. A lookbook should feel cohesive. Pick your light — soft overcast, warm window, hard studio — and use the same description across every clip so the looks belong to the same world.

Generate vertical for social, wide for the site. 9:16 for Reels, TikTok, and Stories where most fashion discovery happens; 16:9 for the campaign film on your product or landing pages.

Run each look several times and cast the best take. Fabric physics and motion are probabilistic — the take where the coat swings just right is often the third or fourth. Generate plenty and select.

Fashion is the use case where AI video has the most to prove and the most to gain: get the fabric and the framing right and you produce a season’s worth of lookbook footage for the cost of a few coffees. When you’d rather start from a tuned prompt than a blank page, LzyPrompt has you covered, and you’ll find more use-case guides on the blog. Generate your first prompt free.

FAQ

How do I get AI video to render fabric movement realistically?

Describe the material and its motion together in the prompt — “structured wool coat that swings as she walks,” “sheer chiffon that floats on the breeze,” “stiff denim that holds its shape.” Naming how the fabric behaves gives the model the physics cues it can’t infer from the garment name alone. Tools strong on motion physics, like Kling, handle this best.

How do I avoid uncanny faces and hands in fashion clips?

Frame for the garment rather than the face. Use tighter crops on fabric and detail, fuller silhouette shots, and three-quarter or rear angles. Since shoppers are evaluating the clothing — fit, drape, texture — garment-led framing both hides the model’s weakest rendering areas and shows exactly what drives the purchase.

Which AI video tool is best for fashion lookbooks?

Kling and Runway Gen-4 lead for fashion — Kling for human motion and fabric physics, Runway for keeping the same model and garment consistent across multiple clips. Sora is strong for editorial camera moves and lifestyle mood, and Veo 3 for photographic realism when a face is in frame. Many lookbooks mix tools per shot.

What aspect ratio should fashion lookbook videos use?

Generate 9:16 vertical for social discovery (Reels, TikTok, Stories), where most fashion shopping now happens, and 16:9 for the campaign film on product pages and landing pages. If you only generate one, vertical is usually the higher-leverage choice for a 2026 fashion audience.

Can AI video replace a full lookbook photoshoot?

For supplementary and social content, increasingly yes — early fashion adopters have cut content spend significantly while increasing output. For the hero campaign imagery and exact-fit accuracy on key pieces, a real shoot still wins. The strongest approach is hybrid: shoot your anchor looks, then use AI video to fill out the long tail of SKUs and social variations.