# GraphRAG SDK

> The most accurate Graph RAG framework. Build knowledge graphs from documents and answer questions over them with graph-based retrieval-augmented generation, on FalkorDB.

## Docs

- [GraphRAG SDK](https://docs.graphrag.falkordb.com/index.md): The most accurate Graph RAG framework. Build knowledge graphs from documents and query them with natural language, on FalkorDB.
- [Getting Started](https://docs.graphrag.falkordb.com/getting-started.md): A step-by-step tutorial for building a knowledge graph from documents and querying it with natural language.
- [Architecture](https://docs.graphrag.falkordb.com/architecture.md): How the strategy-based ingestion and retrieval pipelines fit together.
- [Configuration](https://docs.graphrag.falkordb.com/configuration.md): Comprehensive reference for every configurable component, its parameters, defaults, and usage.
- [Ingestion](https://docs.graphrag.falkordb.com/ingestion.md): How documents become a knowledge graph through the 9-step ingestion pipeline.
- [Extraction](https://docs.graphrag.falkordb.com/extraction.md): How raw text chunks are turned into entities and the relationships between them.
- [Retrieval](https://docs.graphrag.falkordb.com/retrieval.md): How questions get answered using multiple parallel retrieval paths.
- [Storage](https://docs.graphrag.falkordb.com/storage.md): How GraphStore, VectorStore, and EntityDeduplicator persist data in FalkorDB.
- [Graph Schema](https://docs.graphrag.falkordb.com/graph-schema.md): The node types, edges, properties, and indexes that make up the knowledge graph.
- [Ontology Discovery](https://docs.graphrag.falkordb.com/ontology-discovery.md): Bootstrap an ontology from a corpus and propose schema additions as new documents arrive.
- [Ontology Evolution](https://docs.graphrag.falkordb.com/ontology-evolution.md): Safely evolve the schema of an already-populated knowledge graph.
- [Incremental Updates](https://docs.graphrag.falkordb.com/incremental-updates.md): Update, delete, and re-index documents in an existing graph without a full re-ingest.
- [Strategy Reference](https://docs.graphrag.falkordb.com/strategies.md): Every swappable strategy ABC and its built-in implementations.
- [Providers](https://docs.graphrag.falkordb.com/providers.md): Configure LLM and embedder providers, including LiteLLM and OpenRouter.
- [API Reference](https://docs.graphrag.falkordb.com/api-reference.md): Complete reference for every public class and method exported by graphrag_sdk.
- [Benchmark](https://docs.graphrag.falkordb.com/benchmark.md): GraphRAG-SDK results on GraphRAG-Bench, with full methodology and configuration.

## Optional

- [FalkorDB](https://docs.falkordb.com)
- [FalkorDB Cloud](https://docs.falkordb.cloud)
- [FalkorDB Enterprise](https://docs.enterprise.falkordb.com)