ML for 3D Course documentation
Walkthrough
Unit 0. Introduction
Unit 1. What is 3D?
Unit 2. Multi-view diffusion
Unit 3. Gaussian Splatting
Unit 4. Meshes
Unit 5. Capstone
Conclusion
Walkthrough
This section will walk you step-by-step through the easiest way to set up a demo based on LGM, as covered in previous units.
1. Duplicate the model
Start by duplicating the LGM-full model.
- Go to the Repo duplicator.
- Generate a
write tokenin your account settings. Copy it into thetokenfield. - Enter
dylanebert/LGM-fullin thesource_repofield. - Enter
{username}/{model_name}in thedst_repofield, replacing{username}with your Hugging Face account username, and{model_name}with any name you like, such asLGM. - Click Submit.
You will see a link labeled “find your repo here”, which leads to http://huggingface.co/{username}/{model_name}.

Congratulations! You can now use this model with Diffusers, and it will appear in hf.co/models.
2. Duplicate the space
Go to the LGM Tiny space, which provides a simplified image-to-3D demo.
- Click the
Duplicate Spacebutton. - Choose free hardware, or ZeroGPU Nvidia A100 if available.
- Click
Filesin the top right. - Click
app.pyto view the demo source code. - Click
editto change the code. - Replace the two instances of
dylanebert/LGM-fullwith your model path, e.g.{username}/{model_name}. - Click
Commit changes.
Congratulations! You’ve created a demo and met the minimum requirements for this capstone project.
So how do I run it?
The demo requires a GPU, so it won’t work on free hardware. However, there are many free options:
- Run in this notebook: Validate the code quickly.
- Run locally: Clone your space and run it locally.
- Community grant: Building something cool? Apply for a community GPU grant in your space settings.
- Run via API: Less flexible, but runs on free hardware.
The following sections will walk you through each of these options.
Update on GitHub