Arena Leaderboard

Arena Leaderboard

Shows model rankings from lmarena-ai, backed by saved Elo result snapshots and an Arena-Hard-Auto leaderboard file. For people comparing AI models.

Made by lmarena-ai 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: Arena Leaderboard is a Hugging Face Space from lmarena-ai that holds model rankings. Its repository stores a long series of saved Elo result files, one per snapshot, and an Arena-Hard-Auto leaderboard as a CSV file, next to app.py. You can open it in a browser without installing anything.

Who it suits: It suits researchers, model builders and developers who want to see how AI models rank against each other and how the ratings changed between snapshots.

Industry | Who uses it | How they use it | Example
AI research | Researchers | Compare model ratings across saved snapshots | Open the leaderboard and read the current ranking
Model development | Model builders | See where their model stands against others | Look for a model by name in the table
Software | Developers | Pick a model for a project using public rankings | Compare a few candidates by their ratings

If you want to run or test a model yourself, you can skip it, since the page shows rankings and not models.

Where to use it:

• A web browser, on the Space's page on Hugging Face
• A web browser on Windows, macOS or Linux, from a local copy of the files

Use case | Who | What they get back
Read the ranking | Anyone | A leaderboard of models with Elo-style ratings
Follow changes | Researchers | Snapshots saved at different times in the repository
Use the raw data | Analysts | The leaderboard CSV and the Elo result files

How to set it up:

• A web browser, which is all the online page needs
• Python with pip, only if you want the files on your own computer

1. Open the Space's page on Hugging Face to use it online. Nothing needs installing.
2. For a local copy, install the Hugging Face hub package with pip.
3. Download the Space's files into a folder named arena-leaderboard.
4. Move into that folder and open index.html in a web browser, if the folder has one. The record lists app.py as the main file, so check what the folder holds.

pip install -U huggingface_hub
hf download lmarena-ai/arena-leaderboard --repo-type space --local-dir arena-leaderboard
cd arena-leaderboard

How to use it: Open the Space's page and read the leaderboard it shows. The README gives no list of controls, so the page itself shows what you can sort or filter.

Example: Open the page to see how the models rank right now. The repository's saved snapshots hold the earlier states of the same ratings.

For newcomers and experts: Open the Space page and look at the top of the ranking first. An Elo-style rating is a single number that goes up when a model wins comparisons, so a higher number means a better showing.

The repository keeps one Elo result file per snapshot, which makes it a record of how ratings moved over time. It also holds the Arena-Hard-Auto leaderboard as a CSV. The source does not say how the ratings are computed, so read app.py or the project's own notes for the method.

Comparisons between models are collected
Elo ratings are computed from them
Snapshots are saved as result files
The page shows the ranking

Good to know: The README is only a configuration header, so the method and the data sources are not described here.

• License: Apache 2.0
• Cost to run: none named
• Data it reaches: the Elo result files and the leaderboard CSV in its own repository

On Hugging Face, 13 November 2026

• Likes: 5,039
• Built with: static
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/lmarena-ai/arena-leaderboard
• Files: https://huggingface.co/spaces/lmarena-ai/arena-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.