Edit images with AI by description — NetRoom
CAPABILITY

AI image editing by description

Work on a picture you already have: paint over an area with the brush and describe what should change there. The edit applies only to what you painted, the rest of the frame stays as it was, and the seam is blended so no border shows. Leave the mask empty and the description applies to the whole image — that is how a picture is redrawn in a different style.

up to 8
edit zones per run
up to 4
result variants
2048 px
on the long side

How it works

  1. Take a source image

    Upload a file, pick a picture from your gallery or paste a link to an image.

  2. Paint over what should change

    The brush marks an area, the eraser removes stray strokes. There can be up to eight zones, each with its own description of the edit: remove an object in one place and replace the background in another.

  3. Describe the edit and run it

    Write it in ordinary words. One run can return up to four variants; the work happens in the background, so you can switch tabs.

  4. Collect the result

    The finished picture is saved to the gallery as a separate file and the original stays where it was. You can download it from there.

What it does

Input
An image: a file, a picture from the gallery or a link
Edit zones per run
Up to eight, each with its own description
Variants per run
Up to four
Editing without a mask
The description applies to the whole picture — an image-plus-text mode for redrawing in a different style
Result size
Follows the source after downscaling: the long side does not exceed 2048 pixels
Result format
PNG, JPEG or WEBP
Your own photos
Up to four editable shots plus one style reference, files up to 10 MB, images only
Manual tools
Cropping and brush drawing run in the browser, with no model involved and no charge; the result is saved as a separate file
Available in
Studio, the dedicated photo screen, the editor on any attached image, and the chat

What people use it for

Remove an unwanted object from a photo

Paint over the object and describe what should be there instead. Pixels outside the painted area are not recomputed, so faces and details elsewhere in the frame stay intact.

Replace a background

Paint over everything around the subject and describe the new background. The subject itself is untouched, since it never entered the mask.

Several edits in one pass

Up to eight zones with different descriptions: change the colour of clothing in one, remove wires in another, replace the sky in a third.

Redraw a picture in a different style

Without a mask: give the image and a description of the style you want, anywhere from a light touch to a full rework, and the model rebuilds the whole frame.

Models for this task

Full catalog
Image
Google Nano Banana 2 Lite
Lighter Nano Banana 2 image model for faster generation and editing workflows
Free: 3 per day Image $0.042
Image
Seedream 5.0 Flash
Fast, cost-efficient ByteDance Seedream 5.0 tier: image generation and editing at 1K/2K with up to 10 reference images
Free: 3 per day Image $0.023
Image
GPT Image 2
OpenAI image model: precise prompt following, readable text inside images, layout-aware posters, packaging and diagrams
Image $0.08
Image
GPT-Image-2.5 Flare
OpenAI's fastest image model: higher quality than the previous generation at half the latency
Image $0.10
Image
GPT-Image-2.5 Sunburst
OpenAI's most capable image model for complex layouts, reference fidelity and iterative editing
Image $0.18
Image
Grok Imagine Image 2.0
Next-generation Grok image model for text-to-image and prompt-guided editing at 1K and 2K
Image $0.06
Image
Kling IMAGE 3.0
Kling image model: native 2K output, realistic textures and materials, image-to-image from a reference
Image $0.034
Image
Krea 2 Large
Larger Krea 2 variant for rawer, more flexible outputs with stronger photorealism and weighted reference control
Image $0.076
Image
Krea 2 Medium
Faster Krea 2 variant for stable, consistent generation with controllable prompt strength and weighted reference guidance
Image $0.036
Image
Krea 2 Turbo
Krea 2 Turbo — the speed tier of Krea 2: aesthetic images in seconds, style references and moodboards, 8 aspect ratios.
Image $0.052
Image
Nano Banana 2.1
Google's newest Nano Banana: image generation and editing at 1K/2K/4K with up to 14 reference images
Image $0.042
Image
Qwen-Image-3.0
Alibaba image model balancing quality and speed: generation, editing, multilingual typography, up to 3 references
Image $0.036

What else it does

The things people usually discover only once they start working.

  • Mask-based editing is available on some of the models that generate images.
  • How strongly the picture is redrawn is set by the wording of your description.

Frequently asked questions

How do I remove an unwanted object from a photo?

Paint over it with the brush and write what should be in its place. Only the painted area changes, and the seam with the rest of the frame is softly blended, so the edit leaves no visible border.

Can I replace the background without touching the person in the shot?

Yes. Paint over the background around the figure and describe what it should become. Everything left outside the mask comes back from the source image unchanged.

What can I use as a source image?

A file from your device, a picture from your gallery or a link to an image. Up to four of your own shots plus one style reference can go in at once, with a 10 MB limit per file.

What happens to the original picture after an edit?

The result is saved to the gallery as a separate file and the original stays where it was, so you can come back to it at any point and edit it differently.

How is editing billed?

You pay for the edits actually performed, and the price is shown before the run. A zone whose mask has been fully erased is simply skipped and not charged; if a finished result cannot be saved, the money is returned.

Can I edit a picture inside a conversation?

Yes. Ask the model to change the last picture generated in the thread and the edit applies to it. Mask-based work with a precise selection lives in Studio and in the editor for an attached image.

Try it on your own task

A starting balance is already on your account, no card required. You pay only for what you generate.