CortexDB
A RAG-powered memory database with built-in knowledge graphs. Ingest structured documents and conversational prompts with dual pipelines, auto-extract entities & relationships, and query with vector + graph search — powered by any LLM.
Get Started → API ReferenceVector Search
Semantic similarity search on contexts, entities, and history using pgvector
embeddings.
Knowledge Graph
Automatic entity & relation extraction with weighted edges and 1-hop / 2-hop graph
traversal.
Dual Pipelines
SimpleMem for online conversational prompt synthesis, and PageIndexing for hierarchical document processing.
Multi-LLM Support
Gemini, OpenAI, Anthropic, Azure, OpenRouter — switch with a single API call.
Hybrid Search
Combines vector similarity with graph-based context for richer, more accurate results.
Official SDKs
Python, JavaScript/TypeScript, and Java client libraries with full type safety.
Quick Start
# Clone and run with Docker Compose
git clone https://github.com/harshit-sandilya/CortexDB.git
cd CortexDB/memory
docker compose up -d
This spins up PostgreSQL with pgvector and the CortexDB backend on
http://localhost:8080. See the
Getting Started guide to configure an LLM and ingest your first document.
Built With
| Component | Technology |
|---|---|
| Backend | Spring Boot 3.5 · Java 21 |
| Database | PostgreSQL 16 + pgvector |
| AI Framework | Spring AI 1.1 |
| Migrations | Flyway |
| ORM | Hibernate 6 + JPA |
| Containerization | Docker Compose |