Skip to content
Ends Oct 15, 2026, 11:59 PM UTCleftSee pricing
Kamply
DownloadBook a walkthrough

Background removal in Kamply

Background removal that measures its own edge

BiRefNet runs on your Mac's GPU and writes a soft-alpha PNG beside the original.

The longer description

BiRefNet runs on your Mac's GPU, writes a soft-alpha PNG beside the original, and the alpha check reads coverage, halo and edge width, with the halo signed against your brand's canvas colour, before the cutout goes into a post.

Free. Keep the files. Akash replies with times.

Finished work, built in the app.

Every piece here was built in Kamply at the size it ships. Sample brands are labelled as such.

A 4:5 post on a dark field: a fan of year-in-review cards, the front one listing an example year of days logged, meals and streaks in lime numerals
NutriScanYour 2026 is already counted. Feed post · 4:5
A 4:5 post on charcoal headed One studio, every format, over four lime-outlined frames for a post, a reel, a square and a deck
KamplyOne studio, every format. Feed post · 4:5

Sample brands from the app’s library

Darnfold, Boxkin, Woofstead and Latelift are invented brands, so every figure in their work is fictional.

Darnfold, a Kamply sample brand. A square post: a close-up of a torn jeans seam with model road workers, cones and a spool of thread on it, over the line Black Friday road works on your inner thighsample brand
Darnfold · sample brandRoad works on your inner thigh. Feed post · 1:1
Boxkin, a Kamply sample brand. Carousel slides telling a founding story with one cardboard box: an unowned Release notes label, a launch set for Friday, a late sticker, and a handover note taped onsample brand1 / 6
Boxkin · sample brandWe each thought the other had the release notes. Carousel · 4:5
Boxkin, a Kamply sample brand. A square diagram post: a cardboard box with eyes in mid-air between two desks, with three numbered labels explaining who owns it at each pointsample brand
Boxkin · sample brandMid-air, it’s still Ana’s box. Feed post · 1:1
Darnfold, a Kamply sample brand. Carousel slides on dark navy: one pair of jeans with four repairs, then close-ups of the knee patch, the patches above the knee and the oval patch on the upper thighsample brand1 / 5
Darnfold · sample brandBack four times. Carousel · 4:5
Woofstead, a Kamply sample brand. Carousel slides laid out like a script: a late-night chat between a dog owner and a vet nurse, with a photo of a spaniel and stage directions for the dog in italicssample brand1 / 5
Woofstead · sample brandRolo ate a sock at 10:41 pm. Carousel · 4:5
Boxkin, a Kamply sample brand. A landscape hiring post: a handover note taped to a cardboard box reads Wanted, our second product designer, with done and next lines filled insample brand
Boxkin · sample brandWanted: our second product designer. LinkedIn post · 1.91:1
Darnfold, a Kamply sample brand. A 9:16 story for Mothering Sunday: a parcel tag addressed to Mum hangs from a jeans waistband, above a line saying repairs are free and postage is paid both wayssample brand
Darnfold · sample brandA repaired pair, posted to Mum. Story · 9:16
See all 36 pieces

What alpha-check reads before a cutout can go into a post

Coverage0 to 100% of the frame
ships
Below 2%, subject destroyedAbove 98%, nothing was removed
Halo-35 to +70 luminance
0+35, hard fail

Negative means the fringe sinks into your canvas colour, which is what a clean matte usually reads.

Edge band width0 to 8 px
0.6 px6.0 px

Above 6.0 px it only warns: very soft, fine for hair and fur.

Every number on these rails is a constant in the engine's alpha_check.py, where coverage runs 0.02 to 0.98, a halo of +35 hard fails, and the edge band only warns below 0.6 px or above 6.0 px.

What it actually does

Kamply is a Mac desktop app.

What it does, in detail

Matting is not its default route to a cutout. The engine probes the ready-made PNG shelf first, and a matte has to record why the shelf lost. When it is the right call you ask in the embedded terminal, and nobg.sh loads BiRefNet locally, mattes at its 1024px training resolution, and writes name--nobg.png next to the original, which is never touched. About 2 to 6 seconds an image on an Apple M4. Then alpha-check measures the edge before the post gate will accept it.

What this is not

This is not a browser tool.

What it leaves to you

There is no upload box here and no remove background button in the app. The cutout happens on your Mac, in a paid desktop app, on a file already sitting in your project folder.

How it runs, start to finish

Three steps, in the order the app actually does them.

  1. Ask for it in the terminal

    There is no remove background button. You ask in the embedded terminal, and the crew runs nobg.sh on the image already in your project folder. For a person who has to mirror the viewer that is the right call. For an object or a plate of food the engine wants a ready-made cutout instead, and it makes you say why the shelf lost.

  2. BiRefNet mattes it on your Mac

    The model loads locally and runs on the Mac's GPU, about 2 to 6 seconds an image on an M4. The mask is resized back to the source's full width and height and applied with no threshold, so hair and fur keep their soft edge.

  3. The edge gets measured

    alpha-check reports coverage, halo and edge band width, and writes a peek PNG of the cutout on your brand's canvas colour. A halo names its own fix, imgedit.sh dehalo, which erodes the alpha inward and invents no colour.

The numbers, and where they come from

Every figure on this page comes from Kamply's build log or a named published source. Nothing is modelled.

Model
BiRefNet, run locally
Time per image
About 2 to 6 seconds
Output
PNG with 8-bit alpha
Formats in
jpg, jpeg, png, webp, bmp, tif, tiff
Halo hard fail
+35 on a 0 to 255 scale
First run download
About 2 to 3 GB, plus weights
What each number means
Model
ZhengPeng7/BiRefNet through transformers, on your own machine
Time per image
Measured on an Apple M4, 10-core, 16 GB, using the Mac's GPU
Output
Written beside the input as name--nobg.png. The original is never changed
Formats in
Any other still extension is refused by name instead of guessed at. Video files route to the clip path and come back as a ProRes 4444 alpha .mov
Halo hard fail
Signed towards your brand's canvas colour, read from tokens.css
First run download
torch and its dependencies, then 880 MB of BiRefNet weights on top. Never part of the app install

Questions people actually ask

If yours is not here, ask it in the walkthrough. The form below reaches a human.

Can I upload an image on this page and get a cutout back?

No. This site has one backend surface, the walkthrough form, so there is nowhere to send a file. The free tools here do their whole job in your browser, and background removal happens in the Mac app on a file already in your project folder.

Does the image leave my Mac?

No. BiRefNet runs on your machine through transformers, so the matte is computed on your Mac and the image itself is never uploaded anywhere. The only thing that comes over the network is the model: about 880 MB of weights, pulled once and then cached. The hosted version of this model was turned down in the decision record for exactly that reason, because it uploads every client image to a public third-party space. The model does no sampling, so the same file and the same weights give you the same PNG every run.

What does the alpha check catch, and what does it miss?

It hard-fails coverage at or above 98 percent, which means nothing was removed, coverage at or below 2 percent, which means the subject was destroyed, and a halo at or above +35 on a 0 to 255 scale. Edge width only warns. It does not catch segmentation mistakes, so a chopped arm or a scrap of floating debris passes every number. That is why every check writes the peek for you to look at.

Why would the same cutout pass for one brand and fail for another?

The halo is signed towards your canvas colour, read from tokens.css. A dark fringe sinks into a dark field and rings on a light one, so the verdict follows the background the cutout will actually sit on. Grade it against some other background colour and you learn nothing about the post it is headed for.

Can it sharpen a small or low-resolution image?

No. Kamply has no upscaler. The engine's rule runs the other way. Stock photos are fetched at original resolution, because the 1300px default was reading soft in a 1080 wide frame and a board got rejected over three 867x1300 photos. A small source gives you a small cutout.

What do I need to run it?

An Apple silicon Mac on macOS 11 or later. Kamply is a signed, notarized DMG, with no web version, no mobile app, no background removal API and no batch panel. Background removal is not installed with the app. The first run prints one pip install line and exits, then pulls about 2 to 3 GB of torch and 880 MB of weights, once.

Next, if this was not it

Bring one brief to the walkthrough.

We build the first direction with your brand during the walkthrough.

Akash, who built Kamply, replies with times. The walkthrough is free: he builds your first campaign direction on your brand, and you leave with the files.