sourabhmarne777/rag-assistant-ollama
📦 Open Source Projectsourabhmarne777
A local RAG assistant powered by Ollama, Qdrant, and LangChain for private document and web-based Q&A.
The rag-assistant-ollama repository offers a robust framework for deploying local RAG solutions. At its core, the system utilizes LangChain to manage document ingestion and retrieval workflows. By integrating Qdrant, the project ensures efficient semantic search capabilities, allowing the model to retrieve relevant chunks from uploaded PDFs or scraped web pages. The use of Ollama allows users to run the Mistral model locally, ensuring data privacy and reducing latency. The Streamlit-based interface provides a seamless user experience, supporting session-isolated chat history which maintains context throughout the interaction. This tool is highly modular, making it an excellent starting point for developers who want to experiment with local vector databases, document parsing, and LLM-based retrieval systems in a Python-centric environment.
💡Highlights
- ├─Local RAG with Ollama & Mistral
- ├─Qdrant vector DB integration
- └─Streamlit UI with session isolation
🎯For
- ├─AI Developers
- └─Data Engineers