Meldry studio

Take one brief to a finished image.

Choose Create, Face swap, or Mockup. Add the required inputs, then build the prompt, render, compare versions, and refine in the same workspace.

Your ChatGPT session runs every generation.Sign in from the header before generating.

Workflow

Place uploaded artwork on a product surface or supplied scene.

Logo or designArtwork to reproduce exactly

Choose imagePNG, JPEG, or WebP

Scene referenceOptional product and camera cue

Choose imagePNG, JPEG, or WebP

Brand context Required when no scene reference is attached.

Art direction (optional)

Product surfaces (up to 4)

Variations in one image

Visual directionBlend up to 4

Extract a style from an imageSubjects and text are ignored; only visual treatment is saved.

Prompt modelChatGPT

The selected model writes the production prompt. GPT Image 2 renders the final image through the same signed-in session.

Add artwork, choose a product, and provide a scene reference or brand context.

Start with a workflow.

Choose Create, Face swap, or Mockup, then add the required inputs. Meldry builds the prompt and starts the render.