ESEILANE
A high-performance knowledge graph engine built for large language models and GraphRAG, using sparse matrix algebra for sub-millisecond graph traversals, with full OpenCypher query support and Python, JavaScript and Docker access.
ESEILANE is a knowledge graph engine aimed at large language models, GraphRAG and AI-native applications, using sparse-matrix and linear-algebra based query execution to deliver sub-millisecond graph traversals. It stores nodes and relationships as a property graph model with rich attributes and full OpenCypher query support.
It is designed for horizontal scale with a distributed, pay-as-you-grow architecture and zero-overhead multi-tenancy, and for enterprise use with role-based access control, audit logging, and encryption at rest and in transit. Its own published benchmarks, run on an AWS r6i.4xlarge instance, show it completing node lookups, multi-hop traversals, complex Cypher queries, bulk edge inserts and GraphRAG context building faster than Neo4j and ArangoDB on the same hardware.
It runs as a Docker image, with Python and TypeScript client libraries, and its core engine is being rewritten in Rust; a native vector index, a GraphQL API, and a managed cloud offering are listed as still in progress.