How much does Proteus v0.2 cost?

It is billed by GPU time on an L40S rather than a flat per-image price. A typical 20-second generation is about $0.019, so roughly 52 runs per dollar. Heavier settings — more denoising steps, more outputs, larger dimensions — take longer and therefore cost more. Current rates are shown in the Hormouz dashboard.

Can it edit an image or only generate one?

Both. Supply an image for image-to-image and use prompt_strength to control how far the result moves from the original. Supply an image plus a mask for inpainting, where only the white areas of the mask are regenerated and the black areas are preserved.

What is the difference between v0.2 and v0.3?

v0.2 is the general-purpose release with the strongest prompt adherence. v0.3 is the anime update, retuned for anime and illustration work with a different default sampler. Pick v0.2 for photographic and general subjects, v0.3 for anime styling.

How is Proteus v0.2 served on Hormouz?

Behind an asynchronous predictions API: submit a prediction, then poll until the images are ready. Point your base URL at Hormouz and use a Hormouz key.

proteus-v0.2: image generation model

Proteus v0.2 is an SDXL-family image model with notably strong prompt adherence, supporting text-to-image, image-to-image and inpainting at 1024x1024. Served asynchronously through Hormouz from inside Iran.

Pricing:: ~$0.000975 / sec (Nvidia L40S GPU)
Type:: image

Image generation

Proteus v0.2 builds on OpenDalle v1.1 and is tuned for prompt understanding — the model's own notes describe it as surpassing MJ6 on following what you asked for while approaching its stylistic range. Version 0.2 also blends in RealCartoonXL at a very small weight to repair the earlier version's weakness with anime and cartoon tags.

It covers three modes in one model. Text-to-image generates from a prompt alone. Image-to-image reworks a picture you supply, with prompt_strength controlling how far it departs from the original. Inpainting takes an image plus a mask and regenerates only the masked area, leaving the black regions untouched.

Output defaults to 1024x1024 and you can request up to four images in a single run. The sampler is selectable across seven schedulers, with KarrasDPM as the default, and denoising steps run from 1 to 100 — 20 steps for speed, 20 to 35 for more detail.

This model bills by GPU time rather than a flat per-image rate, which means a short run costs less than a long one. A typical generation takes about 20 seconds on an L40S and works out around $0.019. Hormouz provides stable access 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":"datacte/proteus-v0.2","input":{"prompt":"A beautiful sunset"}}'

Specs

Vendor	Datacte
Type	Image
Context window	—
Pricing	~$0.000975 / sec (Nvidia L40S 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":"datacte/proteus-v0.2","input":{"prompt":"A beautiful sunset"}}'

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": "datacte/proteus-v0.2", "input": {"prompt":"A beautiful sunset"}})
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: "datacte/proteus-v0.2", input: {"prompt":"A beautiful sunset"} }),
});
const { id } = await res.json();

Buy the proteus-v0.2 API, paid in rial

Buy credit with an Iranian bank card; no foreign card or account needed.
One Hormouz key reaches proteus-v0.2 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 Proteus v0.2 cost?: It is billed by GPU time on an L40S rather than a flat per-image price. A typical 20-second generation is about $0.019, so roughly 52 runs per dollar. Heavier settings — more denoising steps, more outputs, larger dimensions — take longer and therefore cost more. Current rates are shown in the Hormouz dashboard.
Can it edit an image or only generate one?: Both. Supply an image for image-to-image and use prompt_strength to control how far the result moves from the original. Supply an image plus a mask for inpainting, where only the white areas of the mask are regenerated and the black areas are preserved.
What is the difference between v0.2 and v0.3?: v0.2 is the general-purpose release with the strongest prompt adherence. v0.3 is the anime update, retuned for anime and illustration work with a different default sampler. Pick v0.2 for photographic and general subjects, v0.3 for anime styling.
How is Proteus v0.2 served on Hormouz?: Behind an asynchronous predictions API: submit a prediction, then poll until the images are ready. Point your base URL at Hormouz and use a Hormouz key.

Updated 2026-10-13