TRELLIS

TRELLIS

Creates 3D content from a single image with the microsoft/TRELLIS-image-large model, in a Gradio page you can open in a browser or run on your own computer.

Made by trellis-community 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: TRELLIS is a Gradio app that creates 3D content from an image. It is built on the microsoft/TRELLIS-image-large model and aims at scalable, flexible 3D generation. The Space runs on Hugging Face's ZeroGPU hardware, so you can try it in a browser.

Who it suits: It suits 3D artists, game developers and researchers who want to see what an image-to-3D model produces without setting anything up.

Industry | Who uses it | How they use it | Example
Game development | Indie developers | Turn a concept picture into 3D content for a scene | Try a picture of a prop and look at the 3D result
Design and 3D art | Artists | Start a 3D piece from a reference image | Feed in a sketch-like reference and see a first 3D draft
AI research | Researchers | Try the TRELLIS model before reading its paper | Open the Space and test one image

If you have no images to start from, you can skip it.

Where to use it:

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

Use case | Who | What they get back
Quick 3D draft | Artists and developers | A 3D result made from one image
Model check | Researchers | A first look at how the model handles a test image
Local run | Developers | A copy of the app running with their own Python packages

How to set it up:

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

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 5.34.2, then the packages from requirements.txt.
5. Start the app with python app.py.

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

How to use it: Open the page, or the one your local copy starts, and give the app an image. The app then generates 3D output from it and shows the result in the same page.

Example: Give the app a picture of a single object. It sends the picture to the TRELLIS model and returns a 3D version of that object in the page.

For newcomers and experts: Open the Space page and try one image first. You do not need to install anything, and the result shows what the model can do before you decide to run a copy.

The repository holds app.py next to a trellis package, an extensions folder and a wheels folder, and the app uses the microsoft/TRELLIS-image-large model. The authors describe the method in a paper that the Space page lists. Locally you need to supply the hardware yourself, since the online copy uses ZeroGPU.

You give the app an image
app.py passes it to the TRELLIS model
The model generates 3D output
Gradio shows the result in the page

Good to know: The input is an image, and the output is 3D content. The source gives no limits on image size or output format.

• License: MIT
• Cost to run: no key, account or paid service named, and the online copy runs on Hugging Face's ZeroGPU hardware
• Data it reaches: the images you give it and the microsoft/TRELLIS-image-large model

On Hugging Face, 13 November 2026

• Likes: 817
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/trellis-community/TRELLIS
• Files: https://huggingface.co/spaces/trellis-community/TRELLIS/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.