SAM3 + Qwen3.5

SAM3 + Qwen3.5

A browser demo of multi-object tracking that combines the SAM 3 model with a 2B Qwen3.5 model. Suits computer vision developers and researchers who want to try tracking without installing anything.

Made by prithivMLmods 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 browser demo of multi-object tracking, which means following several things at once as they move. It combines two named models, SAM 3 and a 2B Qwen3.5 model, inside a small Gradio page that runs on Hugging Face hardware.

Who it suits: It suits computer vision developers and researchers who want to see how these two models work together, and hobbyists curious about tracking.

Industry | Who uses it | How they use it | Example
Computer vision | Developers | Try tracking with two named models before building on them | Run a clip and watch objects followed over time
Research | Vision researchers | Look at how the demo pairs SAM 3 with Qwen3.5 | Read app.py to see how the models connect
Video editing | Editors | Explore tracking as a way to follow subjects | Test a short clip with a few moving objects
Education | Students | See multi-object tracking in action | Run the example files in the repository

You can skip it if you need a production tracker with an API and support, since this is a demo.

Where to use it:

• A web browser: open the Space page on Hugging Face
• Windows, macOS or Linux: run a copy with Python
• An MCP client: the Space carries an MCP tag on Hugging Face

Use case | Who | What they get back
Track several objects | Developers | A result that follows each object over time
Try the models quickly | Researchers | A working demo to read and run
Run your own copy | Tinkerers | The same app on your own computer

How to set it up: The online page needs nothing. A local copy needs Python and the packages in requirements.txt.

• A web browser for the online demo
• Python for a local copy
• The packages listed in requirements.txt

1. To use it online, open the Space page on Hugging Face.
2. To run a copy, install the Hugging Face hub package.
3. Download the Space into a local folder.
4. Move into the folder and install Gradio 6.30.0 and the listed packages.
5. Start the app.

pip install -U huggingface_hub
hf download prithivMLmods/SAM3-Plus-Qwen3.5 --repo-type space --local-dir SAM3-Plus-Qwen3.5
cd SAM3-Plus-Qwen3.5
pip install gradio==6.30.0
pip install -r requirements.txt
python app.py

How to use it: Start the app from its folder, or open the Space online. The repository ships example files, so load one first and run it. Then try your own clip to see how the tracking follows the objects.

python app.py

Example: You load one of the example files from the repository and run the demo. The page returns a result in which each detected object is followed across the clip.

For newcomers and experts: If you are new to this, start with a supplied example file and watch what the page marks. A short clip with a few clearly separate objects is the easiest thing to read.

If you know the field, the app builds on the facebook sam3 model and the 2B Qwen3.5 model. The README is only a one-line tagline, so read app.py to see how the two are connected. The hosted Space runs on Hugging Face's zero-a10g hardware.

You load a clip
The Gradio app passes it to the models
SAM 3 and Qwen3.5 find objects
The tracking result appears

Good to know: The README gives only a short tagline, so the page and app.py hold the real details. Both models are downloaded from their own homes on Hugging Face.

• License: apache-2
• Cost to run: none named
• Data it reaches: the files you give it

On Hugging Face, 13 November 2026

• Likes: 23
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/prithivMLmods/SAM3-Plus-Qwen3.5
• Files: https://huggingface.co/spaces/prithivMLmods/SAM3-Plus-Qwen3.5/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.