Open Chinese LLM Leaderboard

Open Chinese LLM Leaderboard

A leaderboard of large language models for Chinese, run by BAAI as a Gradio Space with datasets for requests and results. For researchers and teams comparing models on Chinese tasks.

Made by BAAI 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 leaderboard of large language models for Chinese, run by BAAI. It lists many open models, from Yi-34B to small community models. Its dataset names, such as requests, results and dynamic_model_information, point to a system that tracks evaluation requests and scores. The README is only the Hugging Face configuration pointer.

Who it suits: It suits researchers and engineers choosing a model for Chinese-language work, and model authors who want to see how their model ranks. It also suits developers who want to read how a leaderboard Space is organized.

Industry | Who uses it | How they use it | Example
Research | NLP researchers | Compare open models on Chinese evaluations | Look up where a model sits in the ranking
Software | ML engineers | Shortlist a model for a Chinese product | Compare a few candidates from the list
Model authors | Open-model builders | Check how their release ranks | Find their model in the table
Education | Students | Learn how a leaderboard Space is built | Read app.py and the src folder

Anyone who needs rankings for English-only tasks can skip it, since this board is about Chinese.

Where to use it: The leaderboard runs in two places.

• A web browser, on the page of the Space at Hugging Face
• Windows, macOS or Linux, as a local copy run with Python

Use case | Who | What they get back
Model ranking | Researchers | A table of models with their evaluation results
Model lookup | Engineers | Where a specific model stands against others
Code study | Developers | A repository with the app, a dynamic-update script and source folder

How to set it up: The online page needs nothing installed. A local copy needs the following.

• Python and pip
• Windows, macOS or Linux
• The packages listed in requirements.txt
• Access to the datasets the Space names, which hold requests and results

1. Install Python.
2. Install the Hugging Face hub package and download a copy of the Space.
3. Move into the new folder and install Gradio 4.36.1 and the packages in requirements.txt.
4. Start the app with python app.py.

pip install -U huggingface_hub
hf download BAAI/open_cn_llm_leaderboard --repo-type space --local-dir open_cn_llm_leaderboard
cd open_cn_llm_leaderboard
pip install gradio==4.36.1
pip install -r requirements.txt
python app.py

How to use it: Open the page and read the leaderboard table. The source does not describe the tabs or filters, so follow what the page shows. On a local copy, this command starts the page.

python app.py

Tool | What it does
app.py | Holds the Gradio page
update_dynamic.py | A script in the repository for updating model information
src folder | Holds the app's supporting code
Results datasets | Hold the scores that the leaderboard shows

Example: You open the page and look for a model you are considering. The table shows where it ranks among the listed models, so you can compare it with a few others.

For newcomers and experts: If you are new to this, a leaderboard is a table that ranks models by their scores on the same tests. Open the page, find a model you know, and see what it scores.

If you know the field, the Space reads several datasets named after open-cn-llm-leaderboard: requests, private-requests, results, private-results and dynamic_model_information. It also names a datasets entry from the open-llm-leaderboard project. The repository has a pre-commit config, a Makefile and a pyproject file, and it runs on upgraded CPU hardware.

Models are requested for evaluation
Results are stored in datasets
The app reads the datasets
The leaderboard table shows the scores

Good to know: The README holds no description, so the benchmarks and scoring are not documented in the source. Check the page for the tests it reports.

• License: Apache 2
• Cost to run: none named
• Data it reaches: the leaderboard datasets the Space names
• Hardware: upgraded CPU at Hugging Face

On Hugging Face, 13 November 2026

• Likes: 128
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/BAAI/open_cn_llm_leaderboard
• Files: https://huggingface.co/spaces/BAAI/open_cn_llm_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.