Edit Images with AI

Select any area of an image, describe what you want changed, and the AI transforms just that region. Everything outside the selection stays untouched.

How it works

01

Position the frame

Drag the generation frame over the area you want to change. Everything inside the frame is what the AI will edit.

02

Describe the change

Write what you want in plain language. "Change the sky to sunset" or "replace the chair with a bench". The AI handles the rest.

03

Generate and iterate

Hit generate. Adjust the prompt or reposition the frame if needed. Different models give different results, so experiment.

Features

Frame-based editing

The generation frame defines the edit area. Anything outside stays untouched. Resize and reposition it freely on the canvas.

Natural language prompts

Describe changes in plain English. No masks or manual selection tools needed.

Multiple AI models

Switch between models for different styles. Each handles edits differently.

Layer-based workflow

Every edit creates a new layer. Compare versions, undo changes, or stack multiple edits.

Grid snapping

Frames and images snap to a grid for precise alignment. Everything lines up when you scale back.

Upload your own images

Drag any photo onto the canvas and start editing. No import steps needed.

Common use cases

Replace objects in a scene
Change the style of a specific area
Refine details and fix artifacts
Adapt images for different contexts
Add or remove elements from a composition

In short

Targeted edits without regenerating the whole image. Natural language control, nondestructive layers, and multiple AI models to choose from.