VBench Leaderboard

VBench Leaderboard

The leaderboard page for the VBench project from Vchitect, run as a Gradio app on CPU hardware. For researchers and engineers who want to see how models rank on VBench.

Made by Vchitect 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 the leaderboard page for VBench, a project from Vchitect. It runs as a Gradio app on CPU hardware and names a Vchitect dataset for leaderboard submissions. The README lists three papers by their numbers: 2311.17982, 2411.13503 and 2503.21755.

Who it suits: Researchers and teams that build models and want to see how those models rank on VBench, and readers who follow the papers behind it.

Industry | Who uses it | How they use it | Example
AI research | Model researchers | Check where a model ranks on VBench | Open the Space and read the board
Product teams | ML engineers | Compare candidate models before choosing one | Look up two models on the board
Education | Students | Read a VBench paper next to its leaderboard | Open the Space beside one of the listed papers

People who want to run the benchmark themselves can skip it and use the VBench code repository the README points to.

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
View the board | Researchers | The VBench leaderboard page
Read the code | Developers | The Gradio app in app.py and its settings in constants.py
Follow the papers | Students | Three paper numbers to look up

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 4.36.1 and the packages from requirements.txt.
5. Start the app.

pip install -U huggingface_hub
hf download Vchitect/VBench_Leaderboard --repo-type space --local-dir VBench_Leaderboard
cd VBench_Leaderboard
pip install gradio==4.36.1
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. The Gradio page that comes up is the leaderboard.

python app.py

Example: You open the Space to see where a model sits on the board. If you want the benchmark itself, you follow the README's pointer to the VBench repository.

For newcomers and experts: If you are new, open the Space in your browser and read the board. Nothing needs installing, and the README names the papers if you want the background.

For experts, the whole Space is four files: app.py, constants.py, requirements.txt and the README. It runs on Gradio 4.36.1 with cpu-basic hardware and names the Vchitect/vbench_leaderboard_submission dataset.

You open the Space
Gradio runs app.py
The page shows the leaderboard
You read the ranking

Good to know:

• License: mit
• Cost to run: none named
• Data it reaches: the Vchitect/vbench_leaderboard_submission dataset it names
• Not for: none stated in the README

On Hugging Face, 13 November 2026

• Likes: 365
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/Vchitect/VBench_Leaderboard
• Files: https://huggingface.co/spaces/Vchitect/VBench_Leaderboard/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.