How your mockups are made

Last updated: October 22, 2026

An image model designs and draws the empty rooms; your painting itself is never drawn or changed by it. It's placed into each room by our own code, as you approved it, never repainted.

1. You approve the photo. We straighten it and correct the colour in your browser. You check the corners and size before anything else happens.
2. A language-and-vision model reads the painting (from OpenAI) to choose rooms that suit its colours, mood and subject, and writes a short plan for each room.
3. An image model draws the empty room from that plan, with a blank canvas on the wall at your painting's true size.
4. Our code places your painting into the blank canvas, as approved, never repainted, and checks the result: size, frame and perspective. Rooms that fail are redone or the room goes back to your plan.
5. You get two versions: exact colours, where every pixel of the painting is as you approved it, and room light, where the room's light and shadow fall across it.

What we promise: your painting and photos are never used to train anything, ours or anyone else's. The providers we use don't train on data sent through their API and keep it for up to 30 days only to prevent abuse (OpenAI: your data). Rooms are generated images; if you show a mockup to buyers, you may want to say the room is illustrative.