After AI Studio generates an image, you don't have to start over if something isn't quite right. The built-in editing tools let you refine specific elements — change the background, adjust the garment, swap accessories, or tweak the pose — all from a conversational chat interface without leaving the creation screen.
The Editing Interface
When you open a generated image, you'll see two main areas:
Left panel — the Styling Elements chat, where you describe edits in natural language and interact with the AI
Right panel — your generated image(s), with version tracking, comparison tools, and download options
The top bar shows shortcuts for working with your images: Cmd + hover to compare versions, Shift + Click to recreate, and Click to select an image for editing.
Editing Elements
The core editing tool in AI Studio. You describe what you want to change, and it generates an updated version of your image. There are two ways to start an edit:
Pick an element from the menu
When you click on a generated image, the Edit element panel appears in the chat. Choose the element you want to change, then describe the edit below.
Element | What it edits |
Background | The environment, setting, or backdrop behind the model |
Model | The model's appearance — face, hair, skin tone, expression |
Pose | The model's body position and stance |
Accessories | Jewellery, bags, shoes, hats, and other props |
Garment | The product itself — fit, colour, styling |
Other | Anything else — camera angle, specific details |
Click on a generated image to select it
The Edit element panel appears — pick an element (e.g., Background)
Describe the change in the text field (e.g., "minimalist white studio with soft shadows")
Click Send to generate the edited version
Thinking Mode
When Thinking Mode is enabled, AI Studio may ask clarifying questions before making an edit — especially when your instruction is ambiguous or has multiple possible interpretations.
For example, if you say "add shorts to her outfit", Thinking Mode might ask: "What style of shorts should be added?" and offer options like Denim shorts, Athletic shorts, Cargo shorts, or Biker shorts.
How to toggle it: Look for the brain icon in the bottom bar of the chat panel. When it has a purple dot, Thinking Mode is on.
Turn it on when you want more control and don't mind answering a follow-up question to get a more precise result
Turn it off when you want the AI to just make its best guess and generate immediately
Visual Edit
Visual Edit lets you draw directly on the image to select a specific region you want to change. Instead of describing what to change, you show AI Studio where to change it.
Click the Visual Edits section at the bottom of the chat panel
On the image, you'll see Edit Region at the top with a prompt: "Draw a selection on the image to edit"
Draw a selection around the area you want to modify
Describe what you want changed in that region
Click Send to apply the edit
Visual Edit is especially useful when text alone can't pinpoint the exact area — for example, removing a specific accessory, fixing an awkward shadow in one corner, or adjusting a detail that only appears in part of the image.
Recreate
At the bottom of the Edit element panel, below the six element options, you'll see Recreate — "Start over with same options".
Recreate generates a completely new image using the same settings (model, pose, background, product) but with a fresh result. Use it when the overall settings are right but the specific output didn't land — maybe the pose looks stiff, the lighting fell flat, or the garment drape isn't natural.
Unlike editing, Recreate doesn't modify the existing image. It generates a new one from scratch with identical configuration.
Tip: You can also Shift + Click on any image in the grid to recreate it instantly.
The CREATE Bar
At the bottom of the chat panel, the CREATE bar gives you control over every new generation or edit:
Control | What it does |
Image count (1–5) | How many variations to generate per new creation |
Additional instructions | Optional free-text field for extra guidance |
AI Model (Auto, etc.) | Which generation engine to use |
Resolution (1K, etc.) | Output quality |
Mode icons | Toggle between image generation, video, Thinking Mode, and upload |
Every time you send a text edit, visual edit, or element edit, these settings apply to the output.
Version Tracking
Every edit creates a new version of the image. You can see version numbers in the top bar (e.g., "VERSION 2") and compare them:
Cmd + hover over an image to compare it with the previous version
Click to select an image for further editing
Shift + Click to recreate
Images that have been edited show an EDITED badge in the top corner, so you can always tell which versions are original and which have been modified.
Choosing the Right Edit Method
Scenario | Best method |
You want to change a specific element (background, pose, etc.) | Edit Elements — pick from the menu |
You want to change a specific area of the image | Visual Edit — draw a region and describe the change |
The image isn't working but your settings are right | Recreate — start fresh with the same configuration |
Your edit is ambiguous and you want the AI to ask first | Enable Thinking Mode — get clarifying questions before generation |