SmolVLM

SmolVLM

A Gradio demo from HuggingFaceTB for the SmolVLM2-2.2B-Instruct model, with example images included, that runs online in the browser or on your computer with Python 3.10.

Made by HuggingFaceTB 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: SmolVLM is a Gradio demo that HuggingFaceTB published as a Hugging Face Space. It names the HuggingFaceTB/SmolVLM2-2.2B-Instruct model, and its repository holds a folder of example images beside app.py. At Hugging Face it runs on zero-a10g hardware.

Who it suits: People who want to try the SmolVLM2 model in a browser, and developers who want the demo's files to read or run themselves.

Industry | Who uses it | How they use it | Example
Machine learning | Model testers | Open the Space to try the model it names | Looking at how the model handles the example images
Software development | Gradio developers | Download the files and read app.py | Studying how a model demo is wired up in Gradio
Education | Students | Run a copy with Python 3.10 on their own machine | Following the listed steps from download to launch

Skip it if you want a hosted service with an API, since this is an app you open or run yourself.

Where to use it: SmolVLM runs in these places.

• A web browser: open the Space's page on Hugging Face
• Python 3.10 on Windows, macOS or Linux: run a copy of the Space

Use case | Who | What they get back
Try it online | Anyone with a browser | The Gradio demo running from the Space
Run it locally | Developers with Python 3.10 | A copy of the app started from app.py
Read the code | Learners | The main file, a requirements list and the example images

How to set it up: Online use needs only a browser. A local copy needs a few things first.

• A web browser for the online Space
• Python 3.10 on Windows, macOS or Linux for a local copy
• The Python packages listed in requirements.txt

1. Install the Hugging Face hub library with pip.
2. Download the Space's files into a folder called SmolVLM2.
3. Move into that folder.
4. Install Gradio at the version the Space uses.
5. Install the packages listed in requirements.txt.
6. Start the app with Python.

pip install -U huggingface_hub
hf download HuggingFaceTB/SmolVLM2 --repo-type space --local-dir SmolVLM2
cd SmolVLM2
pip install gradio==5.38.2
pip install -r requirements.txt
python app.py

How to use it: Online, open the Space's page and use the demo in your tab. On your own computer, start it from the folder with the command below.

python app.py

Tool | What it does
python app.py | Starts the Gradio app from the main file
pip install -r requirements.txt | Installs the packages the app needs
hf download | Fetches the Space's files into a local folder

Example: You download the Space with the hf command, install Gradio 5.38.2 and the listed packages, then run python app.py. The demo starts from your local copy of the files.

For newcomers and experts: If you are new, open the Space's page and look at the example images it ships with. Running it online needs no setup, so start there before you try a local copy.

If you know Gradio, the setup is plain: one main file, a requirements list, Python 3.10 and Gradio 5.38.2. A local copy pins that same version. The Space itself uses zero-a10g hardware at Hugging Face, so a local run depends on your own machine.

Open the Space or run app.py
Gradio serves the demo
You use it in your browser

Good to know: It is a single Gradio demo, so what you can do is limited to what its page offers.

• Version: Gradio 5.38.2 and Python 3.10
• Model: HuggingFaceTB/SmolVLM2-2.2B-Instruct
• Size: 18 files in the repository
• License: apache-2
• Cost to run: none named
• Data it reaches: the model it names, none else named
• Not for: no forbidden uses are named

On Hugging Face, 13 November 2026

• Likes: 84
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/HuggingFaceTB/SmolVLM2
• Files: https://huggingface.co/spaces/HuggingFaceTB/SmolVLM2/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.