# Hypabase

> A Python library for storing and querying n-ary relationships with provenance tracking. SQLite-backed, zero configuration.

Hypabase is a Python library for storing and querying relationships between entities. A single edge connects two or more nodes, every edge tracks where it came from (source and confidence), and the whole graph lives in a local SQLite file with no server or configuration. Use it to build knowledge graphs, RAG pipelines, and structured agent memory. Python 3.10+. uv add hypabase.


## Getting Started

- [Home](https://docs.hypabase.app/latest/index.md): Quick install and first hypergraph — go from zero to a working knowledge graph
- [Getting Started](https://docs.hypabase.app/latest/getting-started/index.md): Step-by-step setup with SQLite persistence, namespace isolation, and provenance
- [Concepts](https://docs.hypabase.app/latest/concepts/index.md): Core mental model — hyperedges, provenance, vertex-set index

## Guides

- [Traversal](https://docs.hypabase.app/latest/guides/traversal/index.md): Find neighbors, shortest paths, and multi-hop connections through hyperedges
- [Provenance](https://docs.hypabase.app/latest/guides/provenance/index.md): Track where every relationship came from and how confident you are in it
- [Batch Operations](https://docs.hypabase.app/latest/guides/batch-operations/index.md): Ingest thousands of nodes and edges efficiently with batch, upsert, and cascade delete
- [CLI Quickstart](https://docs.hypabase.app/latest/guides/cli/index.md): Create nodes, edges, and run queries from the terminal without writing Python
- [HIF Import/Export](https://docs.hypabase.app/latest/guides/hif/index.md): Import and export hypergraphs using the HIF interchange format
- [MCP Server](https://docs.hypabase.app/latest/guides/mcp/index.md): Set up the MCP server so AI agents in Claude Code, Claude Desktop, Cursor, or Windsurf can use Hypabase as structured memory

## Examples

- [Medical Knowledge Graph](https://docs.hypabase.app/latest/examples/medical-kg/index.md): Build a clinical knowledge graph with treatment events, provenance, and query patterns
- [RAG Extraction Pipeline](https://docs.hypabase.app/latest/examples/rag-extraction/index.md): Extract structured relationships from documents and query them as a hypergraph
- [Agent Memory](https://docs.hypabase.app/latest/examples/agent-memory/index.md): Give AI agents persistent structured memory that survives across sessions and conversations
- [Hybrid Vector Pattern](https://docs.hypabase.app/latest/examples/hybrid-vector/index.md): Combine Hypabase relationship queries with vector similarity search for hybrid retrieval

## Comparisons

- [vs Neo4j](https://docs.hypabase.app/latest/comparisons/vs-neo4j/index.md): Architectural differences between Hypabase and Neo4j for n-ary relationships
- [vs Vector DBs](https://docs.hypabase.app/latest/comparisons/vs-vector-dbs/index.md): How Hypabase and vector databases complement each other
- [vs Mem0](https://docs.hypabase.app/latest/comparisons/vs-mem0/index.md): Architectural differences between Hypabase and Mem0 for agent memory

## API Reference

- [Client](https://docs.hypabase.app/latest/reference/client/index.md): Complete Hypabase client API — every method with parameters, return types, and examples
- [Models](https://docs.hypabase.app/latest/reference/models/index.md): Pydantic models for nodes, edges, incidences, stats, and validation results
- [CLI Reference](https://docs.hypabase.app/latest/reference/cli/index.md): CLI command reference — init, node, edge, query, stats with all flags and options