llm-colosseum / app.py
PLB's picture
Initial commit
f65f671
raw
history blame
704 Bytes
import gradio as gr
import pandas as pd
# Load the results
elo_df = pd.read_csv("elo-20240326.csv")
text_description = """
# 🤼 LLM Colosseum Leaderboard
LLM Colosseum is a new way to assess the relative performance of LLMs. We have them play Street Fighter III against each other, and we use the results to calculate their Elo ratings.
Watch a demo of LLMs playing Street Fighter III [here](https://youtu.be/Kk8foX3dm2I).
More info in the LLM Colosseum GitHub [repository](https://github.com/OpenGenerativeAI/llm-colosseum).
"""
with gr.Blocks(
title="LLM Colosseum Leaderboard",
) as demo:
gr.Markdown(text_description)
gr.Dataframe(value=elo_df, interactive=False)
demo.launch()