> ## Documentation Index
> Fetch the complete documentation index at: https://pavan-c90e8846.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# RAG Applications for Document Understanding

> Browse 12 RAG projects for document understanding, from basic PDF chat to advanced multi-retrieval pipelines with reranking and contextual retrieval.

The RAG Applications section contains 12 projects that demonstrate retrieve-augmented generation techniques for document understanding, knowledge bases, and multi-source question answering. These projects range from simple PDF chatbots to production-ready pipelines with hybrid search, reranking, and OCR integration.

## RAG Projects Overview

| Project | Focus | Description |
| - | - | - |
| Agentic RAG | Agentic retrieval | RAG implementation with Agno and GPT-5 |
| Agentic RAG with Web Search | Hybrid search | Advanced RAG with CrewAI, Qdrant, and Exa for web + document hybrid search |
| Resume Optimizer | Document enhancement | AI-powered resume optimization and improvement tool |
| LlamaIndex RAG Starter | Getting started | LlamaIndex + Nebius RAG starter template |
| PDF RAG Analyzer | PDF chat | Multi-PDF conversational analysis system |
| Qwen3 RAG Chat | Streamlit UI | PDF chatbot interface built with Streamlit and Qwen3 |
| Chat with Code | Code documents | Conversational code explorer and documentation assistant |
| Gemma3 OCR | OCR + RAG | OCR-based document and image processor using Gemma3 |
| Nvidia Nemotron OCR | OCR + parsing | OCR and image parsing using Nvidia Nemotron-Nano-V2-12b |
| Contextual AI RAG | Enterprise RAG | Managed datastores with quality evaluation for production use |
| Advanced RAG with Reranking | Production pipeline | Contextual retrieval, Qdrant hybrid search, reranking, streaming answers, ingestion, and citations |
| Simple RAG | Quick start | Basic RAG implementation with Nebius for rapid prototyping |

## How to Run a RAG Project

Most RAG projects share a similar setup flow:

<Steps>
  <Step title="Enter the project directory">
    ```bash theme={null}
    cd rag_apps/<project_name>
    ```
  </Step>

  <Step title="Configure your environment and indexing strategy">
    ```bash theme={null}
    cp .env.example .env
    ```

    Many RAG projects require embedding model keys (e.g., OpenAI, Nebius) and a vector database connection (e.g., Qdrant).
  </Step>

  <Step title="Install and run">
    ```bash theme={null}
    pip install -r requirements.txt
    python main.py
    ```

    Some projects use Streamlit for the front end. Run with:

    ```bash theme={null}
    streamlit run app.py
    ```
  </Step>
</Steps>

<Tip>
  If you are building a RAG pipeline for the first time, try the **LlamaIndex RAG Starter** or **Simple RAG** first. They require minimal configuration and clearly illustrate how documents are ingested and queried.
</Tip>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.