Virtual try-on API: what it is and how to add one to your app

Most shoppers know virtual try-on as a button on a product page. Underneath that button, on Corlen and on every other provider in this space, an API is doing the actual image generation. If you are building your own storefront, a mobile app, or a tool that needs try-on as a feature rather than a widget, this is the layer you integrate with directly.

What does a virtual try-on API actually do?

• Takes two images as input: a photo of the person and a photo of the garment, usually a flat lay or a product shot from the catalog.
• Returns a generated image: a result showing the garment rendered on that specific person, not a generic model wearing it.
• Runs as an asynchronous job in most implementations: you submit a request, get a job ID back, then poll a status endpoint until the result is ready, since generation takes real time.
• Accepts fit parameters in the better implementations: garment size, body size, or style options like an open or closed jacket change the output, not just the garment choice.
• Sits behind authentication: an API key scopes usage to your account and its rate limits.

How does the integration actually work?

The pattern is close to any third-party image generation API, with one extra step because the work takes long enough to need a job status check rather than a single blocking call.

1. Get an API key from the provider's developer dashboard.
2. Send a request with the person's photo and the garment image, either uploaded directly or referenced by URL.
3. Poll the job status endpoint until the generation finishes.
4. Retrieve the finished image and display it, or pass it into your own storefront or app.
5. Handle errors and rate limits the way you would for any external API: retries with backoff, and a clear fallback state for the shopper if a request fails.

Corlen's own docs describe exactly this flow: a Create Try-On endpoint for a single garment, a Create Combo Try-On endpoint for a full outfit in one call, and a Generation Status endpoint to poll. The full request and response shapes are in the API documentation.

What to check before picking a virtual try-on API

• Documentation quality: can you get a working request running from the docs alone, without emailing support first.
• Pricing model: per generation, a subscription with included credits, or both, and what happens once you go past your monthly allowance.
• Generation speed: a slow response loses the shopper's attention fast if this is powering a live storefront feature rather than an internal tool.
• Photo handling and privacy: whether the provider stores uploaded photos, for how long, and whether it runs face recognition on them.
• Single garment vs combo support: some APIs only render one item at a time; others accept a top, an outer layer, and a bottom together in a single request.

Corlen's API covers both of the last two cases directly: single-garment and multi-garment combo requests, with no face recognition run on any submitted photo. The same engine also powers Corlen's Shopify integration and an in-store kiosk, so a result generated through the API uses the identical model behind those. For how the image generation itself actually works, see how does virtual try-on work.

If you are evaluating a virtual try-on API for your own product, the fastest way to judge output quality is to look at a generated result directly. Try Corlen on your own photo, or read the API documentation for the exact request and response shapes.