# Background Remover: Free, No Upload, Full Resolution

> Remove the background from any photo with AI that runs in your browser. Full resolution, batches, a brush for touch-ups. Free, no sign-up, nothing uploaded.

https://www.kortex-notebooklm.com/tools/background-remover

[Free tools](/tools)

# Remove the background from a photo

Drop a photo and the AI cuts out the subject in a second or two, at full resolution. It runs on your own device, so the photo is never uploaded.

-   Free, no sign-up
-   Runs in your browser. Your files never leave your device.
-   Updated November 10, 2026

## How do I remove the background from a photo?

Drop the photo above and wait a second or two while the AI finds the subject. Then download it as a transparent PNG or WebP. Or put it on a color, a blur or another picture. The AI model runs in your browser, so the photo is never uploaded. The result keeps the photo’s full resolution.

1.  **Add photos.** Drop them, pick them, or paste one with Ctrl V (⌘ V on a Mac). Many at once is fine.
2.  **Check the cut-out.** Use _Compare_ to slide between the original and the result. If a spot is wrong, switch on the Erase or Restore brush (E, R) and paint over it. Ctrl Z undoes.
3.  **Choose the background and download.** Transparent, a color, a blurred copy of the original or your own picture. One photo downloads on its own. Several come as a .zip.

## What kinds of photo work best?

The model works best on people, pets, products, cars and single objects that stand apart from their background. It is weaker on fine hair against a similar color, glass and other see-through things, thin wires and nets, and scenes with several equal subjects. Where it slips, the brush fixes it in a few seconds. _Clean up edges_ and _Feather_ soften what is left.

## Why is the first photo slow?

The first time, your browser downloads the AI model (about 100 MB) and starts it. It is stored on your device afterwards, so later visits start in seconds, and cutting out works offline. With WebGPU (Chrome, Edge and recent Safari and Firefox on a computer with a graphics card), a photo takes a second or two. Without it, the model runs on your processor and a photo takes tens of seconds. Automatic then uses a smaller, faster version of the model.

## What model is it, and is it free for commercial use?

[BiRefNet](https://github.com/ZhengPeng7/BiRefNet)\-lite by Peng Zheng and colleagues, under the MIT license. Best quality reads the photo at 1024 pixels ([a WebGPU-ready export](https://huggingface.co/jiabins0303/birefnet-lite-1024-webgpu), MIT); Fast reads it at 512 ([another export](https://huggingface.co/studioludens/birefnet-lite-512), MIT). We do not use models with non-commercial licenses.

## How does it compare with remove.bg?

|  | This tool | remove.bg (free) |
| --- | --- | --- |
| Where the photo goes | Stays on your device | Uploaded for processing |
| Free resolution | Your photo’s own, up to 50 megapixels | Up to 0.25 megapixels (about 625 × 400) |
| Sign-up, watermark | None | None for the small preview |
| Batches | Yes, with a .zip out | Paid plans and apps |
| Touch-up brush | Erase and Restore | Erase and Restore |
| Speed | A second or two with WebGPU, after a one-off download | A few seconds, plus the upload |

Checked on 4 November 2026 on remove.bg’s [“Is remove.bg free?”](https://remove.bg/help/a/is-remove-bg-free-) page. remove.bg’s model is trained on far more data than ours and handles hard cases such as fine hair better. If you need that, it is the better tool.

## Questions

Is my photo uploaded anywhere?

No. The AI model runs in your browser, on your graphics card when the browser allows it, and the photo never leaves your device. The tool downloads only the model file (about 100 MB, from Hugging Face, once) and the small program that runs it (from jsDelivr). Your browser keeps both for next time.

Is it really free? What about the resolution?

Free, with no sign-up, no watermark and no daily limit. The cut-out keeps your photo’s own resolution, up to 50 megapixels (for example 8,000 by 6,000 pixels). Larger photos are scaled down to that.

Which AI model does it use, and can I use the results commercially?

BiRefNet-lite, an open model by Peng Zheng and colleagues, released under the MIT license, which allows commercial use. The page names the exact files it downloads. The cut-outs are your pictures. We claim nothing on them.

What does it cut out well, and what does it struggle with?

It does best on people, pets, products, cars and clear objects against a different background. It struggles with fine hair against a similar color, glass and other see-through things, shadows you want to keep, and several overlapping subjects of equal size. The Erase and Restore brush is there for those spots.

Why does the first photo take longer?

The first time, the browser downloads the model (about 100 MB) and starts it. After that it stays on your device, so the next visit starts in seconds. On a computer with WebGPU, a photo then takes a second or two. Without it, the model runs on your processor and takes much longer, so Automatic picks the smaller, faster model for you.

Can I do many photos at once?

Yes. Drop as many as you like (up to 200 at a time). The tool cuts them out one after another. You can fix any of them with the brush. Download all gives you one .zip of full-size files.

Can I put the subject on a new background?

Yes: a plain color, a blurred copy of the original photo, or any picture you choose. Or keep it transparent and save a PNG or WebP. A JPG cannot be transparent, so it is saved on white.

Does it work offline?

Yes, once the model has been downloaded. With the page open, cutting out needs no connection.

## More free tools

-   [**White Background Photo** Put product photos on pure white for a shop or marketplace.](/tools/white-background-photo)
-   [**Blur Photo Background** Keep the subject sharp and blur everything behind it.](/tools/blur-photo-background)
-   [**Image to Text (OCR)** Copy the text from photos, scans and screenshots. 100+ languages.](/tools/image-to-text)
-   [**Image Compressor** Shrink JPG, PNG, WebP and AVIF images, one or hundreds at a time.](/tools/image-compressor)
-   [**Merge PDF** Join PDFs into one file. Reorder and rotate pages in your browser.](/tools/merge-pdf)

## Save what you find in one click.

Kortex is a Chrome extension for Gemini Notebook (formerly NotebookLM). It sends AI chats, web pages, posts and YouTube videos into the notebook you pick. It keeps your notebooks organized. It exports them back out.

[Add to Chrome](https://chromewebstore.google.com/detail/kortex/hdapplggdhndkblofffknpmnnnnbncbn)

No card needed.
