Virtual Try-On, Explained
Turn any garment photo into a photorealistic on-model image. This guide covers how the studio works, what each setting does, and how to get the best results from every run.
Why Brands Are Switching to Virtual Try-On
Traditional photoshoots weren't built for today's pace. Here is what changes when the model shot becomes something you generate instead of produce.
A fraction of photoshoot costs
A model shoot costs thousands once you add talent, studio, and retouching — multiplied across every product. A try-on run costs a few credits.
Seconds, not weeks
Booking, shooting, retouching, delivery — traditional production takes weeks per drop. A garment photo becomes a publishable on-model image in under a minute.
Every model, every market
Show the same garment on different looks, body types, and ages without casting a single person, and match your imagery to each market you sell in.
Scales with your catalog
New products don't need new shoots. Hundreds of SKUs are just more runs, not more logistics — your imagery keeps up with your catalog.
How Try-On Works
Every try-on needs two things: the garments you want to show and the model who wears them. Everything else is optional fine-tuning.
Upload your garments
Add up to 5 garment images. Flat lays, ghost-mannequin shots, and on-model photos all work. Combine tops, bottoms, shoes, and accessories to build a complete outfit in a single run.
Pick your model
Choose from built-in AI models across different looks, body types, and ages, or upload your own photo. For your own uploads, use a full-body shot of a single person facing the camera — the clearer the photo, the more realistic the result.
Choose settings and run
Pick the AI model, resolution, and aspect ratio, optionally describe the scene with a prompt, and press Run. The exact credit cost is shown on the button, and your result lands in History within seconds.
Before & After
Every example below started as a single garment photo — no model, no set, no shoot. This is the exact transformation the studio performs on every run.
What Can You Try On?
The studio handles full outfits, not just single garments. Add up to 5 pieces per run and combine categories to style a complete look.
Tops
Shirts, tees, blouses, and knitwear. A flat lay on a plain background keeps prints and necklines sharp.
Bottoms & denim
Trousers, skirts, and jeans. Lay them flat and straight so the cut and leg shape stay accurate.
Dresses
From slip dresses to gowns. Ghost-mannequin shots work best — they show the AI how the silhouette falls.
Outerwear
Coats, jackets, and blazers. Shoot them closed or open, exactly the way you want them worn in the result.
Shoes
Sneakers, heels, and boots. Add them as a separate image alongside the outfit to complete the look.
Accessories
Bags, hats, and jewelry. Combine them with garments in one run to style the whole outfit at once.
What Are the AI Models?
The model picker decides which AI engine generates your image. Both understand garments, fabric, and fit — they differ in fidelity, speed, and credit cost.
Best quality
Nano Banana Pro
Google's flagship generation model. The highest garment fidelity, the most realistic fabric texture and lighting, and support for 1K, 2K, and 4K output. Use it for final assets that go on product pages, ads, and campaigns.
Fast & affordable
Nano Banana
Google's standard generation model. Faster and cheaper per run, with great everyday quality. Use it to explore outfits, test garments, and iterate before committing credits to a final render.
A good workflow: draft with Nano Banana until the outfit and scene feel right, then re-run the winner with Nano Banana Pro at a higher resolution.
Choosing an Aspect Ratio
The aspect ratio sets the shape of your image. Pick it based on where the image will live — resizing afterwards crops your model, so it pays to decide upfront.
Auto
Matches your model photo's original framing
1:1
Product cards and marketplace listings
4:5
Instagram feed posts and lookbooks
9:16
Stories, Reels, and TikTok
16:9
Website banners and hero sections
Picking a Resolution
Resolution controls the pixel size of the final image. Higher resolutions cost more credits, so match the resolution to the destination instead of defaulting to the maximum.
1K
Drafts, social posts, and fast iteration. Sharp enough for most on-screen use.
2K
Product pages and ads where customers zoom in on fabric and details.
4K
Print, billboards, and campaign hero shots that need maximum detail.
Writing Prompts
The prompt is optional — without one, the model simply wears your garments in the original setting. Use a prompt when you want to control the scene: background, lighting, pose, mood, or photography style.
A prompt directs everything around the garment, not the garment itself. The AI always stays faithful to the clothing you uploaded, so describe the world you want it photographed in.
"Dress the model with the dark charcoal double-breasted wool coat from the reference image. She is standing on a New York street corner beside a black street lamp, urban city background, cinematic street fashion photography, natural daylight, realistic fabric texture."
- Name the garment you uploaded so the AI anchors on it: "the uploaded brown blazer".
- Describe the scene: location, background, time of day.
- Set the light: "natural daylight", "soft studio lighting", "golden hour".
- Add a style: "editorial fashion photography", "e-commerce catalog shot".
Common Mistakes to Avoid
Most disappointing results trace back to the input images, not the AI. These are the patterns we see most often — and what to do instead.
Use a cropped or group photo as the model
Headshots, waist-up crops, and photos with multiple people confuse the AI about who wears the garment and how it should fit.
Use a full-body photo of one person
A single person, visible head to toe, facing the camera. This gives the AI the full silhouette it needs for a realistic fit.
Upload blurry or cluttered garment shots
Low-resolution photos, busy backgrounds, and folded garments hide the details the AI needs, so texture and cut get lost in the result.
Use sharp images on a clean background
A well-lit flat lay or ghost-mannequin shot on a plain background preserves fabric, print, and stitching in the final image.
- Want a different background? Ask for it in the prompt — the model photo's setting is kept by default.
- Garment details looking soft? Re-run with Nano Banana Pro at 2K or higher.
- Layering multiple garments? Add each piece as a separate image instead of one combined photo.
What to Do With Your Results
A try-on image is usually the start of the pipeline, not the end. Every result in your History can flow straight into the other studios.
- Image Upscale — scale your result up to 4K for print and campaign use.
- Background Changer — swap the scene without re-running the try-on.
- Reframe — extend the image to a new aspect ratio for banners and stories.
- Image Generation Studio — restage the scene, change the pose, or turn your result into new campaign visuals.
- Video Studio — animate your on-model image into a short fashion video.
- UGC Try-On Video — turn a product and model into a ready-to-post UGC-style clip.
Flat Lay, Ghost Mannequin, On-Model — What's the Difference?
Flat lay
A garment photographed from above on a flat surface. The simplest input for virtual try-on — a phone photo on a clean background is enough.
Ghost mannequin
A garment shot on a mannequin that is removed in post, leaving its 3D shape. A great input because it already shows how the garment holds its form.
On-model photo
A garment worn by a person. It is the output the studio generates — and also a valid input when you restyle an existing product photo.
Frequently Asked Questions
Ready to see your products on a model?
Sign up free and run your first virtual try-on in under a minute.
Get Started