Docs-Talk 🤖📚

Unlock Knowledge Instantly with AI-Powered Insights

---

🔧 Built With

✨ Powered by: LangChain, Ollama, Vector Embeddings, Streamlit, and local LLMs for private, powerful, and customizable AI without relying on external APIs.

---

🌟 Why SIMPLE_RAG_CHAT?

Tired of manually digging through long documents?
SIMPLE_RAG_CHAT transforms static PDFs into dynamic knowledge bases. Whether you're researching, studying, or analyzing reports, this tool helps you:

• 🔍 Find relevant info fast using semantic search powered by vector embeddings.
• 💬 Chat naturally with your documents like talking to a smart assistant — backed by Ollama's local LLMs.
• 📄 Process PDFs on the fly with automatic text extraction and chunking.
• 🔄 Maintain conversation history across messages for coherent follow-ups.
• ⚙️ Easy to customize & extend — perfect for developers building privacy-first, AI-powered tools.
• 🛡️ Run entirely offline using Ollama — no data leaves your machine!

Ideal for:

• Researchers 🎓
• Developers 👨‍💻
• Students 📚
• Analysts 📊
• Privacy-conscious users 🔐

---

📦 Getting Started

✅ Prerequisites

Before you begin, ensure you have:

• Python 3.9+ (required by pip and modern packages)
• pip – Python’s package installer
• git – for cloning the repository
• Ollama – installed and running locally

💡 Make sure Ollama is running! Start it and pull a model like:

ollama pull deepseek-r1:1.5b

---

🛠️ Installation

1. Clone the repository

   git clone https://github.com/ashishkumar0724/simple_rag_chat 
   

2. Navigate to the project directory:

   ❯ cd simple_rag_chat
   

3. Install the dependencies:

Using pip:

❯ pip install -r requirements.txt

▶️ Run

Run the Streamlit app:

streamlit run main.py

Maintained by Ashish Kumar

⬆ Back to top