InstantMesh

InstantMesh

Turns a single image into a 3D model, and its page says this takes about ten seconds. It comes from TencentARC and suits 3D artists, game developers and researchers.

Made by TencentARC 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: InstantMesh turns one picture into a 3D model. The Space's page says it takes about ten seconds. It is a demo from TencentARC and it runs on ZeroGPU hardware at Hugging Face.

Who it suits: 3D artists, game developers and researchers who want a quick first 3D shape from a single image. It also suits hobbyists who want to try image-to-3D with nothing to install.

Title: Who makes 3D models from one image
Industry | Who uses it | How they use it | Example
Games | Indie developers | Get a rough 3D object from concept art | Turn a character sketch into a first mesh
Design | 3D artists | Start a model from a reference image | Begin a prop from a single photo
Research | Graphics researchers | Try a published image-to-3D method | Test the model on their own pictures
Hobby | Makers and 3D printing fans | Turn a picture into a shape to refine | Convert an image of a toy into a 3D model

If you need a production-ready mesh with clean topology, you can skip it, since this gives a quick first shape.

Where to use it:

• A web browser: the Space runs on Hugging Face with nothing to install
• Windows, macOS or Linux with Python, from a local copy

Use case | Who | What they get back
Turn an image into 3D | Artists | A 3D model made from your picture
Get a quick base shape | Game developers | A first model to refine in your own tools
Try the method | Researchers | A working demo of the InstantMesh approach

How to set it up: Using the Space online needs nothing. A local copy needs Python and the packages the repository lists. The Space runs on ZeroGPU at Hugging Face, so a local copy may need a capable GPU.

• Python with pip
• The Python packages in requirements.txt

1. Install the Hugging Face hub package and download the Space into a folder.
2. Move into that folder.
3. Install Gradio 5.50.0, then the packages from requirements.txt.
4. Start the app with Python.

pip install -U huggingface_hub
hf download TencentARC/InstantMesh --repo-type space --local-dir InstantMesh
cd InstantMesh
pip install gradio==5.50.0
pip install -r requirements.txt
python app.py

How to use it: Open the Space in your browser, or start the local copy with the command below and open the page Gradio serves. Give it one image and wait for the 3D model it makes.

python app.py

Example: You give it a clear picture of a single object on a plain background. The Space builds a 3D model of that object from the one image and shows it to you.

For newcomers and experts: If you are new to this, open the Space and try the example images from the repository first. Then try a picture of your own with one clear object in it.

If you know the field, the Space names sudo-ai/zero123plus-v1.2, facebook/dino-vitb16 and TencentARC/InstantMesh. The repository holds a zero123plus folder, configs and example images. The trade is speed over detail: a model made from one view has to guess the sides it cannot see.

You give it one image
Multiple views of the object are inferred
A 3D model is built from them
The result appears in the page

Good to know:

• Hardware: the Space runs on zero-a10g at Hugging Face
• Input: a single image is enough
• License: apache-2
• Cost to run: none named

On Hugging Face, 13 November 2026

• Likes: 1,609
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/TencentARC/InstantMesh
• Files: https://huggingface.co/spaces/TencentARC/InstantMesh/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.