Retrieval-Augmented Generation (RAG) in Python

This folder contains Python examples for different RAG patterns, including local RAG, agentic RAG, and context stuffing.

Requirements

• Python 3.8+
• API keys for OpenAI (if using OpenAI examples)
• Ollama (if using local RAG examples)

Examples

• context-stuffing-system-prompt/: Injecting context into the system prompt.
• context-stuffing-user-prompt/: Injecting context into the user prompt.
• rag-openai/: Basic RAG using OpenAI and TF-IDF.
• rag-hugging-face/: RAG using Hugging Face models and local retrieval.
• agentic-rag/: RAG using a ReAct agent.
• cag/: Cache-augmented generation using KV cache.
• local-rag/: Fully local RAG setup.
• ragflow-basic/: Basic example using the RAGFlow Python SDK.
• vectorless-rag-pageindex/: Vectorless RAG with PageIndex.

Running the examples

Each example is in its own folder and contains a README.md with instructions on how to run it.