RAG examples

These examples show how to build Retrieval-Augmented Generation (RAG) applications with AkasicDB. Each example opens in Google Colab so you can run it directly.

Hybrid RAG

This example shows how AkasicDB handles vector operations and graph operations together in a single hybrid RAG workflow. It retrieves semantically similar context with vector search, then uses graph relationships to expand and refine the context before generation.

Run Hybrid RAG in Google Colab

LangChain RAG

This example uses LangChain's VectorStore interface and langchain-akasicdb to retrieve documents stored in AkasicDB. It adapts OpenAI's Build a Retrieval Augmented Generation (RAG) App: Part 1 tutorial for AkasicDB.

Run LangChain RAG in Google Colab

RAG without LangChain

Some RAG pipelines need direct control over data access and model execution. This example builds a RAG application that accesses AkasicDB directly without using LangChain. It runs a text embedding model and an LLM directly without relying on external APIs.

Run RAG without LangChain in Google Colab