USO FLUX

USO FLUX

A Gradio app from bytedance-research on ZeroGPU hardware, built around the USO model and three FLUX.1 models its record names. The README gives no description, so the page itself shows what it does.

Made by bytedance-research 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: This Space is a Gradio app from the bytedance-research group, running on ZeroGPU hardware at Hugging Face. Its record names three FLUX.1 models, dev, Krea-dev and schnell, together with the USO model and two image encoders, SigLIP and CLIP. The README is only Hugging Face's default template and gives no description, so the page itself is the best guide to what it does.

Who it suits: It suits people who want to try a FLUX.1 based app from this group, and developers who want to read how it is put together.

Industry | Who uses it | How they use it | Example
Software | Developers | Read the app.py and uso folder to see how the named models are wired | A developer opens the code after trying the page
Research | Machine learning researchers | Try the USO model next to the FLUX.1 models the app names | A researcher runs the page and notes how it behaves
Design | Creators | Explore the page's controls to see what it can make | A creator tries the page and compares the results

You can skip it if you need written documentation, since the README has no usage notes.

Where to use it: You use it in these places.

• A web browser, on the Space's page on Hugging Face
• A local copy on Windows, macOS or Linux, which needs Python

Use case | Who | What they get back
Try the app online | Anyone | Whatever the page's controls produce on screen
Study the code | Developers | The app.py file and the uso folder to read

How to set it up: Online use needs only a browser. A local copy needs Python and the packages the repository lists.

• A web browser, for the online page
• Python with pip, for a local copy
• Python packages from requirements.txt

1. Open the Space's page on Hugging Face to use it online, with nothing to install.
2. For a local copy, install the Hugging Face Hub package.
3. Download the Space's files into a folder and move into it.
4. Install Gradio 5.49.1 and the packages in requirements.txt.
5. Start the app.

pip install -U huggingface_hub
hf download bytedance-research/USO --repo-type space --local-dir USO
cd USO
pip install gradio==5.49.1
pip install -r requirements.txt
python app.py

How to use it: Open the Space's page and use the controls it shows. For a local copy, start it with the command below and open the page it serves in your browser.

python app.py

Example: You open the Space's page and run it with the controls it shows. The result appears on the page, and what it produces depends on those controls, which the source does not describe.

For newcomers and experts: Gradio is a way to put a model behind a simple web page. Start by opening the Space's page in your browser and trying the controls one at a time.

Under the hood, the app is built with Gradio 5.49.1 and runs on zero-a10g hardware at Hugging Face. The record names FLUX.1-dev, FLUX.1-Krea-dev and FLUX.1-schnell from Black Forest Labs, the USO model, google/siglip-so400m-patch14-384 and openai/clip-vit-large-patch14. The source does not say how these fit together.

You open the Space in a browser
You use the page's controls
The app runs on ZeroGPU hardware
The result shows on the page

Good to know: The source is thin, so treat this page as a map of the files and models and check the Space itself for the rest.

• License: apache-2.
• Cost to run: none named. The Space uses zero-a10g hardware at Hugging Face.
• Data it reaches: the models the record names, which are fetched from Hugging Face, and whatever you give the page.

On Hugging Face, 13 November 2026

• Likes: 302
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/bytedance-research/USO
• Files: https://huggingface.co/spaces/bytedance-research/USO/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.