Local-RAG: A Retrieval-Augmented Generation App Using Local Resources

Local-RAG is a Retrieval-Augmented Generation (RAG) application designed to work entirely with local resources.

🚀 Key Components

• Vector Database: Qdrant
• Language Model: Ollama (Llama 3.1)
• Embedding Model: Snowflake/snowflake-arctic-embed-m-v1.5
• User Interface: Streamlit

Planned Enhancements

1. Contextual Conversations
2. History Tracking using SQlite
3. Advanced RAG Features (reranking, query processing, handling complex queries etc)
4. Configurable Setup
5. Better error handling and logging

🛠️ Installation

Step-by-Step Setup

1. Clone the repository

   git clone https://github.com/pmgautam/local-rag.git
   cd local-rag

2. Create virtual environment

   python -m venv venv
   source venv/bin/activate  # Linux/Mac
   # or
   .\venv\Scripts\activate  # Windows

3. Install dependencies

   pip install -r requirements.txt

4. Install and start Qdrant

   # Start Qdrant with persistent storage and default ports
   docker run -d \
     -p 6333:6333 \
     -p 6334:6334 \
     -v $(pwd)/qdrant_storage:/qdrant/storage:z \
     qdrant/qdrant

5. Install Ollama and download model

   # Install Ollama from ollama.com
   # Pull the Llama 3.1 model
   ollama pull llama3.1

📚 Usage

Indexing Documents

# Index documents with default settings
python -m app.indexer --folder /path/to/documents --collection my_docs

Starting the Application

# Start with default configuration
streamlit run app/chat_app.py

# Start with custom configuration
streamlit run app/chat_app.py -- --config path/to/config.yaml

🤝 Contributing

Contributions are welcome! Please send a PR for any improvements you would want to make.