metadata

title: Arabic Chatbot Arena
emoji: 🏆🇦🇪
colorFrom: green
colorTo: blue
sdk: gradio
sdk_version: 4.28.3
app_file: app.py
pinned: false
license: mit
tags:
  - arabic
  - nlp
  - chatbot
  - language-model
  - comparison

Arabic Chatbot Arena

Overview

Welcome to the Arabic Chatbot Arena! This interactive space allows you to compare and evaluate different Arabic language models side-by-side. Test various chatbots, analyze their responses, and contribute to the advancement of Arabic NLP.

Features

Model Selection: Choose from a curated list of state-of-the-art Arabic language models.
Side-by-Side Comparison: Evaluate two models simultaneously for easy comparison.
Customizable Prompts: Use your own system prompts and user inputs to test specific scenarios.
Real-Time Generation: Watch as the models generate responses in real-time.
Adjustable Parameters: Fine-tune generation settings like temperature, top-p, and max tokens.
RTL Support: Full right-to-left text support for a natural Arabic writing experience.

How to Use

Select two models from the dropdown menus.
(Optional) Customize the system prompt to set the context.
Enter your question or prompt in Arabic.
Click "Generate" to see both models respond.
Compare the outputs and use the "Like" buttons to rate responses.
Adjust generation parameters as needed for different results.

Models

This space features a diverse selection of Arabic language models, including:

MohamedRashad/Arabic-Orpo-Llama-3-8B-Instruct
silma-ai/SILMA-9B-Instruct-v1.0
inceptionai/jais-adapted-7b-chat
... (and many more)

Contributing

Your feedback is valuable! If you encounter any issues or have suggestions for improvement, please open an issue on the GitHub repository.

License

This project is licensed under the Apache 2.0 License. See the LICENSE file for details.

Acknowledgements

Special thanks to all the model creators and the Arabic NLP community for their contributions to advancing Arabic language technology.

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference