NitroFusion2

NitroFusion2

Makes images from text in just one or two steps with NitroFusion2, a model built on Z-Image-Turbo, in a Gradio page on ZeroGPU hardware. For people who want quick text-to-image results.

Made by ChenDY on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: NitroFusion2 is a Gradio demo of a text-to-image model that generates a picture in one or two steps. It builds on Z-Image-Turbo. Its research idea is adding adversarial loss to the part of the network that handles text. It runs on ZeroGPU hardware, so you can try it in a browser.

Who it suits: It suits designers, creators and researchers who want to see how fast a distilled text-to-image model can be, and developers who study how such models are tuned.

Industry | Who uses it | How they use it | Example
Content creation | Creators | Get quick images from a text prompt | Type a short prompt and see the picture appear
AI research | Researchers | Look at a one-step or two-step image model | Compare one step with two steps
Software | Developers | Read the demo code for inference and prompts | Open app.py and inference.py

If you need to edit existing pictures, you can skip it, since this is a text-to-image demo.

Where to use it:

• A web browser, on the Space's page on Hugging Face
• A local copy of the files, run on Python 3.12 under Windows, macOS or Linux

Use case | Who | What they get back
Quick image drafts | Creators | A picture in one or two steps
Model comparison | Researchers | A look at a few-step model next to others
Study the code | Developers | Short Python files that call the model

How to set it up:

• A web browser, which is all the online page needs
• Python 3.12 and the packages in requirements.txt, for a local copy

1. Open the Space's page on Hugging Face to use it online. Nothing needs installing.
2. For a local copy, install the Hugging Face hub package and download the Space's files.
3. Move into the downloaded folder.
4. Install Gradio 6.15.1, then the packages from requirements.txt.
5. Start the app with python app.py.

pip install -U huggingface_hub
hf download ChenDY/NitroFusion2 --repo-type space --local-dir NitroFusion2
cd NitroFusion2
pip install gradio==6.15.1
pip install -r requirements.txt
python app.py

How to use it: Open the page, or the one your local copy starts, write a text prompt and generate. The README lists no controls, so follow what the page shows.

Example: Type a short description of a scene and run the demo. The page returns the picture after only one or two model steps.

For newcomers and experts: Open the Space page and run one short prompt. The speed is the point, because a normal image model needs many more steps.

The model, ChenDY/NitroFusion2, sits on top of Tongyi-MAI/Z-Image-Turbo. The repository holds app.py, inference.py and two prompt helpers, prompt_check.py and prompt_edit.py. The README is only a title and a short description, so the details of the method are in the authors' paper.

You write a text prompt
The prompt is checked and edited
NitroFusion2 generates in one or two steps
The page shows the picture

Good to know: The README gives no limits or settings, so those details are not documented here.

• License: Apache 2.0
• Cost to run: none named, and the online copy runs on ZeroGPU hardware
• Data it reaches: the prompts you type and the two models the Space names

On Hugging Face, 13 November 2026

• Likes: 5
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/ChenDY/NitroFusion2
• Files: https://huggingface.co/spaces/ChenDY/NitroFusion2/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.