LLMChat is a minimalist application designed to test and interact with your LLM in a user-friendly way. Seamlessly integrate local and GitHub-based knowledge to enhance your AI's contextual capabilities. 🌟

[🎥 Demonstration]

• Interactive Interface: Use LLMChat like ChatGPT but tailored to your specific knowledge base. 💬
• Custom Knowledge Sources: Link local folders and GitHub repositories to create a dynamic, up-to-date context for your LLM. 📂
• Privacy-Friendly: Runs locally, ensuring complete control over your data. 🔒

To simplify deployment, you can use Docker Compose to run both the frontend and backend.

• Install Docker and Docker Compose.
• Ensure that Ollama is running locally on your machine and accessible at http://localhost:11434 (default configuration).

• Clone the repository:

git clone https://github.com/Bessouat40/LLMChat.git
cd LLMChat

• Start the application with Docker Compose:

docker-compose up --build

The application will be accessible at:

• Frontend: http://localhost:3000
• Backend API: http://localhost:8000

1. Install dependencies and start the backend:

python -m pip install -r api_example/requirements.txt
python api_example/main.py

2. Install dependencies and start the frontend:

npm i && npm run start

LLMChat leverages RAGLight to index and process knowledge bases, making them available for your LLM to query. It supports:

• GitHub repositories 🧑‍💻
• Local folders with PDFs, code, and more 📄

• Setting Up a Pipeline:

from raglight.rag.simple_rag_api import RAGPipeline
from raglight.models.data_source_model import FolderSource, GitHubSource

pipeline = RAGPipeline(knowledge_base=[
    FolderSource(path="<path to folder>/knowledge_base"),
    GitHubSource(url="https://github.com/Bessouat40/RAGLight")
], model_name="llama3")

pipeline.build()
response = pipeline.generate("What is LLMChat and how does it work?")
print(response)

You can find an API example in the api_example/main.py file. This shows how the backend handles requests and interacts with the LLM.

🚀 Get started with LLMChat today and enhance your LLM with custom knowledge bases!