Frequently asked questions

VirtuFit provides the infrastructure for AI virtual try-on: a REST API and a native MCP server that turn one garment image and one shopper photo into a photorealistic on-body preview. We supply the engine; you supply the storefront or the product built on top of it.

If you run Shopify or WooCommerce, our first-party connectors let you switch try-on on without theme code. If you build your own product or agent flow, the API and MCP server are designed for developers and ship with sandbox keys, SDKs, and clear schemas.

Most renders complete in a few seconds through our asynchronous job model. You receive a job ID immediately and a signed webhook when the image is ready, so your interface never blocks while a render is in progress.

Requests declare a garment type of upper, lower, or full_set, so one integration covers tops, bottoms, dresses, jumpsuits, and full outfits. Fabric, drape, seams, and color are preserved from your original product photography.

One product image (a flat-lay, packshot, or existing on-model shot) and one shopper photo. Our validation step checks both before a render is charged, so unusable inputs are caught before they cost you credits.

Every plan includes a monthly render allowance, and additional renders are billed per unit at your plan's rate, ranging from $0.009 down to $0.005 as you scale. Only successful renders are counted, so spikes are predictable rather than surprising.

Yes. The widget SDK is white-label: it inherits your colors, typography, and language, and the fitting room carries your brand rather than ours. Branded modeling presets can further tune the look to a single brand.

The connector and the API both emit try-on events that feed add-to-cart and completed-order attribution. On Growth and Scale, these events are available in the dashboard and through a queryable event stream.

Any MCP-compatible client can call VirtuFit tools for try-on, garment validation, and size guidance. This lets agent platforms and internal copilots offer try-on as a tool call without bespoke integration code.

Shopper photos are processed only to produce the requested render and are deleted automatically after the render window closes. They are never used to train models, and your customers' images are never shared with third parties.

Failed or rejected jobs are not charged. Growth and Scale plans include our quality review channel, where a renderer can be re-run after a model or preset adjustment, with a resolution target of one business day.

Yes. Plans upgrade at any time and are prorated, so a store can begin on Starter, run a live test, and move to Growth or Scale as volumes and requirements grow. Annual billing reduces any plan by 20%.