Why is the amount reserved larger than the amount charged?

The real cost is not known until the job finishes, so Hormouz reserves one cent when you submit the prediction. Once it completes, the charge is computed from the actual GPU seconds used and the difference is released back to your balance.

Is the output transparent?

Yes. The default output is a PNG with an alpha channel. Setting format to jpg loses the transparency, since jpg has no alpha channel. For a solid or blurred backdrop instead, set background_type to a value such as white, green or blur.

Can I supply my own image as the background?

No. The upstream description mentions a "path to an image", but background_type is a plain string field rather than a file input, so the provider never downloads it — the path it refers to is one inside the model's own container. A URL there reaches the model as literal text and the run fails. Use a keyword such as white or blur, or an RGB triple written as a string like "[0,255,0]", and composite a custom backdrop yourself onto the transparent PNG.

background-remover: image generation model

Background Remover cuts the subject out of a photo and returns it on a transparent background, built on the InSPyReNet segmentation model. Served asynchronously from inside Iran through Hormouz.

Pricing:: ~$0.000225 / sec (Nvidia T4 GPU)
Type:: image

Image generation

Background Remover, by 851 Labs, isolates the main subject of an image and removes everything else. The output is a PNG with an alpha channel, so edges stay soft and the cut-out drops straight onto any background. Common uses are e-commerce product shots, ID and profile photos, and prepping images for design work.

It is served through Hormouz behind an asynchronous predictions API. Submit a prediction with your image, then poll until the result is ready. The required input is image; optional controls are threshold, which hardens the edge (default 0, meaning a soft alpha), reverse, which removes the subject instead of the background, background_type, which swaps in a solid colour or a blur, and format for the output. Keep format at png to preserve transparency — jpg has no alpha channel. The input image is mirrored into our storage before we forward it, so a file hosted inside Iran works.

Pricing is metered on GPU time on an Nvidia T4 at $0.000225 per second. A typical run takes two to three seconds, so about $0.0005 an image. Hormouz reserves one cent when the job is submitted and releases the difference once it finishes, so the amount reserved is larger than the amount charged.

Hormouz provides stable access to Background Remover from inside Iran, with no foreign credit card or circumvention tooling required.

Quickstart

# Create the prediction, then poll until succeeded
curl -s https://api.hormouz.net/v1/predictions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"version":"851-labs/background-remover","input":{"image":"https://example.com/input.jpg"}}'

Specs

Vendor	851 Labs
Type	Image
Context window	—
Pricing	~$0.000225 / sec (Nvidia T4 GPU)
API	predictions

Code example

cURL

# Create the prediction, then poll until succeeded
curl -s https://api.hormouz.net/v1/predictions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"version":"851-labs/background-remover","input":{"image":"https://example.com/input.jpg"}}'

Python

import requests
BASE = "https://api.hormouz.net/v1"
H = {"Authorization": "Bearer ng-YOUR_API_KEY", "Content-Type": "application/json"}
r = requests.post(f"{BASE}/predictions", headers=H,
    json={"version": "851-labs/background-remover", "input": {"image":"https://example.com/input.jpg"}})
print(r.json()["id"])

JavaScript

const BASE = "https://api.hormouz.net/v1";
const res = await fetch(`${BASE}/predictions`, {
  method: "POST",
  headers: { Authorization: "Bearer ng-YOUR_API_KEY", "Content-Type": "application/json" },
  body: JSON.stringify({ version: "851-labs/background-remover", input: {"image":"https://example.com/input.jpg"} }),
});
const { id } = await res.json();

Buy the background-remover API, paid in rial

Buy credit with an Iranian bank card; no foreign card or account needed.
One Hormouz key reaches background-remover and 600+ other models, and you pay only for what you use.
No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.

Create an account

Use via Hormouz

Connect to this model and 600+ others with one API key, then see usage down to each request.

Request company access

FAQ

How much does Background Remover cost?: Billing is metered on GPU time on an Nvidia T4 at $0.000225 per second. A typical run takes two to three seconds, so about $0.0005 an image. Current rates are shown in the Hormouz dashboard.
Why is the amount reserved larger than the amount charged?: The real cost is not known until the job finishes, so Hormouz reserves one cent when you submit the prediction. Once it completes, the charge is computed from the actual GPU seconds used and the difference is released back to your balance.
Is the output transparent?: Yes. The default output is a PNG with an alpha channel. Setting format to jpg loses the transparency, since jpg has no alpha channel. For a solid or blurred backdrop instead, set background_type to a value such as white, green or blur.
Can I supply my own image as the background?: No. The upstream description mentions a "path to an image", but background_type is a plain string field rather than a file input, so the provider never downloads it — the path it refers to is one inside the model's own container. A URL there reaches the model as literal text and the run fails. Use a keyword such as white or blur, or an RGB triple written as a string like "[0,255,0]", and composite a custom backdrop yourself onto the transparent PNG.

Updated 2026-10-13