TripoSplat

TripoSplat

Runs the TripoSplat model from VAST-AI in a browser demo with a custom page and built-in example inputs. The Space has no written description, so the model page is the place to see what it produces.

Made by VAST-AI 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 demo of the TripoSplat model from VAST-AI. It also names the BiRefNet model. The page is a custom index.html, the repository holds example inputs, and the demo runs on zero-a10g hardware at Hugging Face. The README has no description of what the model outputs, so this page sticks to what the repository shows.

Who it suits: It suits people who follow VAST-AI's models and want to try TripoSplat in a browser, and developers who want to read how a Gradio app with a custom page is put together.

Industry | Who uses it | How they use it | Example
AI research | Researchers | Try a new model from a known lab | Open the Space and run one of the example inputs
Software | Developers | Read a Gradio demo with a custom page | Look at app.py, model.py and triposplat.py
Creative tech | Curious makers | Test a model before reading its paper | Run the demo and look at the output

You can skip it if you need a written guide, since the source gives no usage notes.

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

• A browser: open the Space's page on Hugging Face and nothing needs installing.
• Windows, macOS or Linux: run your own copy with Python 3.12.

Use case | Who | What they get back
Trying the model | Anyone curious | The model's output for an example or your own input
Reading the code | Developers | A small Gradio app with model code in separate files

How to set it up: You need these first for a local copy:

• Python 3.12 and pip
• A machine able to run the models, since the Space uses ZeroGPU

The online page needs none of this. For your own copy, follow the steps below.

1. Install the Hugging Face tools.
2. Download the Space's files into a folder.
3. Move into that folder.
4. Install Gradio at the version the Space uses, then the other packages.
5. Start the app.

pip install -U huggingface_hub
hf download VAST-AI/TripoSplat --repo-type space --local-dir TripoSplat
cd TripoSplat
pip install gradio==6.15.2
pip install -r requirements.txt
python app.py

How to use it: Open the Space in a browser, or start your own copy with the command below. Then use the page to pick or load an input and run the model.

python app.py

Example: You open the Space, choose one of the built-in example inputs and run the model. The page shows what the model returns for that input. The source does not say what form the output takes.

For newcomers and experts: If you are new to this, open the Space and start with an example input before you try your own. Read the model page on Hugging Face to learn what TripoSplat is meant to produce.

If you know the field, the repository is small. app.py serves the page, model.py and triposplat.py hold the model code, and example_inputs_b64.py stores example inputs as base64 text. The Space names VAST-AI/TripoSplat and ZhengPeng7/BiRefNet as its models.

Open the Space page
Choose or load an input
The model runs on a GPU
The page shows the output

Good to know: The source describes the Space only through its files and run steps. Check the model page for what the model does and how it may be used.

• License: mit
• Cost to run: none named
• Data it reaches: the inputs you load, and model files downloaded from Hugging Face

On Hugging Face, 13 November 2026

• Likes: 355
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/VAST-AI/TripoSplat
• Files: https://huggingface.co/spaces/VAST-AI/TripoSplat/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.