AI Packshots: Create Packshot in 3 Steps

Every product page needs clean packshots. Ghost mannequins, flat-lays, white-background cutouts: every retailer has a different spec, and reshooting the same garment four ways for every SKU is slow and expensive. Create Packshot solves that.

With On-Model's Create Packshot feature, you upload raw garment photos (phone snaps, hanger shots, or shop-floor pictures) and get back clean, brand-ready packshots in any of four styles. No reshoot, no retouching, no studio time.

This guide walks you through the entire process in the On-Model app, from uploading your first garment to downloading your finished AI packshots.

What you'll need

• An On-Model account. Sign up free if you haven't already.
• Garment photos. Anything that clearly shows the product works: front and back hanger shots, mannequin photos, even phone snaps.
• A few minutes.

Step 1: Upload your garment photos

The Create Packshot wizard starts with the upload step. Drag and drop, click to browse, or pick from your existing assets. You can add 1 to 10 photos of the same garment: front, back, side, detail crops, anything that helps the AI understand the product.

Here's the input pair we'll use for this guide. Two casual phone photos of a tailored blue blazer on a bedsheet, the kind of "raw" snap any showroom or shop floor can produce in seconds:

Step 2: Pick a packshot style

This is where you tell On-Model what kind of packshots you want. Create Packshot ships with four packshot styles, each tuned for a different surface in your catalog:

• Flat-lay: top-down on a clean surface, perfect for editorial product grids and lookbooks.
• Ghost mannequin: invisible mannequin, garment hovers in 3D, the standard for most fashion PDPs.
• Marketing-ready: editorial scene with props, lighting, and atmosphere, ready for campaigns.
• White cutout: pure white background with sharp edges, ideal for marketplaces and feed images.

Pick the style that matches the surface you're producing for. If you sell on multiple channels, stack a few instructions in a single job. One ghost-mannequin set for your PDPs, one white-cutout set for marketplace feeds, one marketing-ready set for the homepage banner. They all run from the same input photos.

You'll also configure the basics:

• Background, surface, composition, props: text or reference image.
• Framing and angle: front, three-quarter, back, top-down, eye-level, low-angle, and so on.
• Output size: 1K, 2K, or 4K resolution.
• Aspect ratio: 1:1, 3:4, 4:3, 9:16, or 16:9.
• Format: PNG or JPG.
• Variations: generate 1 to 4 variations per instruction.

For this guide we used: ghost mannequin style, white background, 2K resolution, 3:4 aspect ratio, 3 variations.

Step 3: Review and start processing

The final step shows a summary of everything you've configured:

• Garment photos: all the inputs you uploaded.
• Instructions: the packshot styles and settings you stacked.
• Total outputs: how many images will be generated.
• Credit cost: how many credits this job will use (3 credits per image at 1K, 5 at 2K, 10 at 4K).

Give your project a name and hit Start Processing. You'll be redirected to the Jobs page where you can track progress in real time. Most packshot jobs finish in 1 to 3 minutes.

Your ghost-mannequin packshots

Once processing completes, your finished packshots are waiting in the job results. Here are the three ghost-mannequin variations the AI generated from our two phone photos: front, three-quarter, and back, all on a clean white studio background.

Every packshot reconstructs the same garment. Colors, textures, fabric weight, and the printed inner lining are preserved across all three angles. Drop them straight into your PDP carousel.

Same garment, different style: a flat-lay variation

Need a flat-lay version for your editorial grid or product feed? Add a second instruction to the same job. From the exact same input photos, the AI produces a clean top-down flat-lay packshot with the same blazer, the same fabric drape, and the same lining detail visible:

Same garment, completely different packshot style, zero extra uploads. This is the real win of stacking instructions in a single job: every surface in your catalog covered from one set of inputs.

Tips for sharper AI packshots

1. Upload more angles. Front and back is the minimum; front, back, and side gives the AI more to work with and noticeably sharper results.

2. Show the garment clearly. Hanger shots, mannequin shots, even flat-lays work as inputs. The clearer the product, the cleaner the packshot.

3. Match the style to the surface. Ghost mannequin for PDPs, white cutout for marketplaces, flat-lay for editorial, marketing-ready for hero banners. Different surfaces, different specs.

4. Stack instructions in one job. Need ghost mannequin and white cutout and a campaign shot from the same garment? Add three instructions and run them in one batch.

5. Save your favorites as presets. Once you've dialed in the look for a brand or category, save it. Future jobs become a one-click load.

6. Use 4K for hero placements. 1K is fine for thumbnails and feeds; bump to 4K for landing pages and printed materials.

What's next

You've just generated your first AI packshots from raw garment photos. From here, you can:

• Try Flat-to-Model. Need on-model imagery from the same flat-lay? Turn flat-lay photos into on-model shots with our other AI feature.
• Try Model Swap. Already have on-model photos? Swap the model while keeping the garment pixel-perfect.
• Scale up. Submit batch jobs with hundreds of garments at once.
• Integrate via API. Connect Create Packshot directly to your product catalog pipeline.

Ready to start? Head to the On-Model app and try it with your own products.