AI Chatbot is a conversational application powered by multiple trending Language Learning Models (LLMs)
and enhanced with a Retrieval-Augmented Generation (RAG) feature to chat with your PDFs and documents seamlessly
You can choose from a variety of state-of-the-art LLMs and leverage RAG to bring contextual knowledge into every conversation
Currently supported models include:
llama2-7b
llama3-8b-instruct
claude-3.7-sonnet
deepseek-r1
llama2-70b
Built using Streamlit and Replicate API, the app allows real-time interaction with various open-source LLMs.
This application is deployed on Streamlit Cloud Click to see⎾
- Create a free Replicate account
- Get your Replicate API key from your profile
- [Optional] Set up a local environment if testing locally
git clone https://github.com/MatthieuGillieron/chatbot.git
cd chatbot
pip install -r requirements.txt
streamlit run src/app.py
Once running:
- Paste your Replicate API key in the settings tab
- Choose your model and temperature
- Start chatting!
- Real-time chat with LLMs
- Select among multiple powerful models
- Adjust temperature for creative or precise responses
- Clean and responsive Streamlit UI
Licensed under the MIT License
If you like the project, give it a ⭐ on GitHub!
Feel free to fork, improve, or submit pull requests.