Definition — Generating images of a specific garment or product being worn by a model without a physical photoshoot. Used in fashion e-commerce to show items on-body across looks, scenes and variants.
Virtual try-on takes a product reference — typically a flat-lay or packshot of the garment — and generates images of a persona wearing it, preserving the item's cut, color and key details while the model, pose and scene are synthetic. It is a specialized relative of image-to-image generation: the garment anchors, the generation dresses.
The e-commerce case is straightforward economics: on-body imagery converts better than flat-lays, but a traditional on-model shoot has to be booked, shot and retouched per season and per variant. Generated try-on imagery refreshes a catalog — new colorways, new scenes, seasonal restyling — without another shoot. The fashion e-commerce use case shows the workflow around a consistent brand persona.
Honest limits: generated try-on is illustrative styling, not a fit or sizing tool — it shows how a garment looks in context, not how size M drapes on a specific body. Fine details are the failure mode to review for: intricate patterns, small logos and text on fabric can render approximately, so check hero shots at full size before publishing.
Generating a new image using an existing image as input alongside a text prompt. The input image anchors composition, pose or subject while the prompt steers what changes.
A UGC ad format in which the persona holds, uses or presents the product on camera. The framing pairs a human face with visible proof of use — a staple of social product advertising.
See it live with 100 free credits
No card required. Generate with showcase models on the free tier, then train your own persona on any paid plan.