Choose an image
A clear subject with complete edges is easier to edit. Upload a still JPG, PNG, or WebP image.
AI image editor
Change an image background to white, studio, or tabletop
Upload an existing JPG, PNG, or WebP and describe the new background: white, a studio, or a tabletop. Compare the original with the edited result. You can also adjust colors or try another style without signing up.
Pick a starting point, then adjust the instruction
The original ratio is matched to the closest ratio supported by the model.
When you start an edit, the image and instructions are sent to the AI service. Privacy policy
Your image appears here
Upload an image, then choose how to edit it.
Edits can also change subject details. Check faces, product labels, and text before use.
Three steps from original to result
Describe the change
Say what to change and which details to preserve. Presets give you a starting point; you can edit the instruction.
Review and download
Switch between the original and the result. Use the result for another edit, or restore the original to start over.
What can you change after uploading?
Put a product image on a white background
Choose Background and the White preset. Describe the product shape, colors, and labels you want to keep. This creates a white background, not a transparent cutout. Compare the output with the source before using it to represent a real product.
Swap a gray background for a studio or tabletop
Start with Studio or Tabletop, then edit the color, lighting, and setting in the instruction. The mug example below shows a gray background changed to pale blue and also reveals changes to the mug and shadow.
Turn an image into anime or watercolor art
Choose Style and try the Anime or Watercolor preset. The model regenerates the whole image, so faces, lettering, and product details may change. Review them before downloading.
See the mug background editing example →
Before you edit
How is image editing different from text-to-image generation?
Text-to-image generation starts with a prompt. Image editing starts with an existing image and an instruction describing what to change. You can also select “Continue editing” on a generated image to bring it into this editor.
Which editing mode should I choose?
Choose Background to change the setting, Free edit to adjust colors or lighting, and Style to try anime, watercolor, or illustration. Each mode generates a result from the whole image and your instruction. Presets fill in text that you can change.
How can I describe a change while preserving the subject?
Describe the change first, then name the details to preserve. For example: “Replace the background with a pale blue studio and soft lighting. Preserve the off-white ceramic cup’s shape, handle, color, and position as closely as possible.” Start with one main change and review it before editing again. Labels, lettering, faces, and other subject details can still change.
Which files are supported? Will the dimensions stay the same?
Upload a still JPG, PNG, or WebP up to 3 MB and 25 megapixels. Images are resized when needed before processing. Output uses the model’s 1K size setting. “Closest to original” selects the nearest supported aspect ratio, so output dimensions and proportions may differ from the input.
Can I remove a background to transparency, use a mask, or upscale?
The current editor uses text instructions. It does not offer brush masks, transparent background removal, or professional super-resolution upscaling. The White preset creates a white background, not a transparent one.
Do I need an account? Are my edits retained?
No registration or login is required. Uploaded images and undo history stay in the current page’s memory; refreshing, leaving the page, or switching languages clears them. Download results before leaving. Starting an edit sends the image and instructions to the AI service. The site stores generation records including instructions, result links, aspect ratio, and available IP information; see the privacy policy for details.