raphaelmansuy/edgequake

Health v2 · maintenance only

Adoption not part of the score

2078 stars · 243 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

EdgeQuake is a high-performance Graph-RAG framework written in Rust, inspired by LightRAG, that transforms documents (PDFs, markdown, text) into knowledge graphs for retrieval-augmented generation. It ships as a full self-hosted stack with a REST API, Web UI, PostgreSQL-backed graph (AGE) and vector (pgvector) storage, and six retrieval modes.

Use cases

• build a graphrag pipeline over my pdf documents
• extract entities into a knowledge graph for better rag answers
• self-host a lightrag alternative without python
• query documents with hybrid vector and graph retrieval
• run multi-tenant rag with source traceability
• integrate a graph-based retriever into langchain or open webui
• ingest multi-column pdfs into a searchable knowledge base

When to choose

• you want Graph-RAG quality with Rust performance instead of Python pipeline fragility
• you need entity relationships, lineage, and provenance in retrieval results
• you want self-hosting with PostgreSQL, pgvector, and Apache AGE
• you need multi-tenancy, ingestion cancellation, and production operational features
• you prefer Docker quickstart deployment with OpenAI or Ollama providers

When to avoid

• you only need simple vector similarity search without graph structure
• you cannot run PostgreSQL or Docker in your environment
• you need a lightweight embedded library rather than a full service stack
• your team depends on a mature ecosystem with long-term stability guarantees
• you want a fully managed cloud service

Facets

service · maturity active

rag search-engine vector-database nlp machine-learning pdf api-framework http-server large-language-models artificial-intelligence pdf self-hosted rust self-hosted windows cross-platform graphrag knowledge-graph lightrag-inspired entity-extraction pgvector postgresql-age hybrid-retrieval multi-tenancy pdf-ingestion openai ollama langchain-integration open-webui retrieval-augmented-generation search natural-language-processing docker web-server linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="raphaelmansuy/edgequake")