LGM-Mini

LGM-Mini

Runs a small Gradio demo built on the LGM model and several multi-view diffusion models, on a Hugging Face GPU. It suits people interested in 3D generation research who want to try or read the code.

Made by dylanebert 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: LGM-Mini is a Gradio demo that names four models, ashawkey/LGM, dylanebert/LGM, dylanebert/imagedream and dylanebert/multi-view-diffusion, and one dataset called iso3d. The Space has no description of its own. Going by those names, it works on 3D content made from pictures. It runs on Hugging Face's zero-a10g GPU setting and the repository holds five files.

Who it suits: It suits people who follow 3D generation research and want to try the demo, and developers who want to read a short app that wires several models together.

Industry | Who uses it | How they use it | Example
AI research | Researchers and students | Try a multi-view model demo without installing it | A student opens the Space and runs the demo.
Graphics | 3D artists | See what a model can make from a picture | An artist tries the demo with a picture of a single object.
Software | Developers | Read how a short app uses several models | A developer opens app.py to see which models it loads.

If you need a documented tool with known outputs, skip it, because the Space gives no description of what it returns.

Where to use it:

• A web browser, through the Space's page on Hugging Face
• A local copy on Windows, macOS or Linux, using Python and a capable GPU

Use case | Who | What they get back
Try the demo online | Anyone | The Gradio page, running on Hugging Face's GPU
Run it locally | Developers | The same app on your own computer, if it can run these models
Read the code | Learners | A short app.py that loads the named models

How to set it up:

• Python and pip on Windows, macOS or Linux
• The packages listed in requirements.txt
• A GPU, since the hosted Space uses zero-a10g hardware

1. Install the huggingface_hub package.
2. Download the Space into a local folder named LGM-mini.
3. Move into that folder.
4. Install gradio at version 5.49.1.
5. Install the packages listed in requirements.txt.
6. Start the app.

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

How to use it: Open the Space's page, or start the app locally with the command below. Then use the Gradio page that opens.

python app.py

Tool | What it does
app.py | The main file, which starts the Gradio app
requirements.txt | Lists the Python packages the app needs

Example: You run the download commands and then python app.py. The Gradio page opens in your browser, and you use it with a picture of your choice.

For newcomers and experts: If you are new, open the Space's page and try it with one simple picture. Expect to read the page itself to learn what the controls do.

If you know the field, the model list mixes LGM with ImageDream and a multi-view diffusion model, which fits a pipeline from a picture to several views and then to 3D. That is a reading of the names, not a statement in the source.

Open the Space or run app.py
Give the page an input
The named models run on a GPU
The page shows the result

Good to know:

• The Space has no description, so what it returns is not documented
• The hosted version uses a GPU, so a local run needs capable hardware
• License: mit
• Cost to run: none named
• Data it reaches: the inputs you give the page, and the model files it names

On Hugging Face, 13 November 2026

• Likes: 123
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/dylanebert/LGM-mini
• Files: https://huggingface.co/spaces/dylanebert/LGM-mini/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.