Hi3DGen

Hi3DGen

Generates detailed 3D geometry from one image, using the Stable-X trellis-normal model. A Gradio app for 3D artists, game developers and researchers, running in the browser on Hugging Face.

Made by Stable-X 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 is a Gradio app that builds a 3D shape from a single picture. Its page describes the output as high-fidelity geometry, and it names the Stable-X trellis-normal-v0-1 model. It runs on zero-a10g hardware at Hugging Face, so you can try it in a browser with nothing to install.

Who it suits: 3D artists and game developers who want a starting shape from a reference picture, and researchers who follow single-image 3D work.

Industry | Who uses it | How they use it | Example
Games | Game artists | Get a base shape from a reference picture | Turn a concept image of a prop into a mesh to refine
3D art | Modelers | Try single-image geometry before sculpting by hand | Compare the result with their own model
Research | Graphics researchers | Try the Stable-X model on their own images | Run several pictures and inspect the shapes
Software | Developers | Read app.py to see how the pipeline is wired into Gradio | Reuse the layout for another model

Skip it if you need a full multi-view scan, since the app works from one image.

Where to use it:

• A web browser, on the Space's page at Hugging Face
• Windows, macOS or Linux with Python, as a local copy

Use case | Who | What they get back
Make 3D geometry from a picture | Artists and developers | A 3D shape built from the image
Try the model locally | Researchers | A running copy of the app

How to set it up:

• Nothing at all if you only use the online Space
• Python and pip, for a local copy
• The Python packages listed in requirements.txt

1. Open the Space's page at Hugging Face to use it online.
2. For a local copy, install the Hugging Face hub package and download the Space.
3. Move into the new folder and install Gradio at the pinned version.
4. Install the packages from requirements.txt.
5. Start the app with Python.

pip install -U huggingface_hub
hf download Stable-X/Hi3DGen --repo-type space --local-dir Hi3DGen
cd Hi3DGen
pip install gradio==5.34.2
pip install -r requirements.txt
python app.py

The repository is large, at 251 files, and includes folders named trellis, extensions and wheels. The page does not say what a local run needs beyond the packages.

How to use it: Online, open the Space and add one image through the form on its page. On your own machine, run the app file and open the page Gradio serves in your browser. The page lists no other commands.

Tool | What it does
python app.py | Starts the Gradio app on your computer

Example: You open the Space and upload a clear picture of a single object, such as a chair. The app builds a 3D shape from that one view, and you look it over to judge whether it is a good base for your own work.

For newcomers and experts: If you are new, open the Space online and try one clear picture of a single object on a plain background. Single-view 3D is a guess about the unseen sides, so check the result from several angles.

If you know the field, the app wraps the Stable-X trellis-normal-v0-1 model. The page has a Paper heading with nothing under it, so look for the method elsewhere before relying on details such as normal handling or output format.

Open the Space in a browser
Add a single image
The trellis-normal model builds geometry
You inspect the 3D result

Good to know:

• License: mit
• Cost to run: none named, though the Space runs on zero-a10g hardware at Hugging Face
• Data it reaches: the image you upload, and the model files it pulls from Hugging Face

On Hugging Face, 13 November 2026

• Likes: 716
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/Stable-X/Hi3DGen
• Files: https://huggingface.co/spaces/Stable-X/Hi3DGen/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.