SeedVR2 3B Image Upscaler

SeedVR2 3B Image Upscaler

Restores and upscales a single image in one diffusion step with the SeedVR2-3B model, in the browser on Hugging Face or from a local copy. Suits anyone with a soft or small picture to clean up.

Made by Upsampler 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 restores and upscales images with SeedVR2-3B, a one-step diffusion model from ByteDance-Seed. You give it a picture and it returns a cleaner, sharper version. It serves the image path only, so video is not part of it.

Who it suits: Anyone with a small, soft or noisy picture who wants a better copy without setting up a model first. Photographers, designers and shop owners often need this kind of cleanup.

Industry | Who uses it | How they use it | Example
Photography | Photographers | Sharpen small or soft shots | Run a low-resolution portrait through the Space
Design | Designers | Prepare artwork for a larger size | Upscale a mockup before a print layout
E-commerce | Shop owners | Clean up supplier pictures | Restore a blurry product photo
Archives | Hobbyists | Improve old scans | Run an old family scan through it

People who only need a plain resize with no restoration can skip it.

Where to use it:

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

Use case | Who | What they get back
Restore a soft photo | Photographers | A cleaner, sharper version of the image
Enlarge artwork | Designers | A larger picture to place in a layout
Rescue an old scan | Hobbyists | A restored copy of the scan

How to set it up:

• Python with pip on Windows, macOS or Linux
• The Python packages listed in requirements.txt
• Nothing at all if you only use the Space in a browser

1. Install the Hugging Face hub package.
2. Download the Space's files into a folder.
3. Move into that folder.
4. Install Gradio 5.49.1 and the packages from requirements.txt.
5. Start the app.

pip install -U huggingface_hub
hf download Upsampler/seedvr2-3b --repo-type space --local-dir seedvr2-3b
cd seedvr2-3b
pip install gradio==5.49.1
pip install -r requirements.txt
python app.py

How to use it: Open the Space page in a browser, or start your own copy with the command below. Gradio then serves a small web page where you add an image and run the restoration.

python app.py

Example: You take a small, blurry product photo and add it on the page. The app runs the model and gives you back a restored version of the same picture.

For newcomers and experts: If you are new, open the Space in your browser and try one small photo first. Nothing needs installing for that. A local copy only makes sense once you want to see how the app works.

If you know the field, the model is SeedVR2-3B from ByteDance-Seed, which restores in a single diffusion step instead of many. The app is a Gradio front end in app.py, and the header of that file says what was changed from the reference code.

You add an image
Gradio passes it to app.py
SeedVR2-3B restores it in one step
You get the restored image

Good to know:

• License: Apache-2.0, as the README states
• Cost to run: none named
• Data it reaches: the images you give it, and the SeedVR2-3B model it names
• Not for: video, because this Space serves the image path only

On Hugging Face, 13 November 2026

• Likes: 4
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/Upsampler/seedvr2-3b
• Files: https://huggingface.co/spaces/Upsampler/seedvr2-3b/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.