helix-db
HelixDB is an OLTP graph database with native vector and full-text search built in Rust on Object Storage.
What is helix-db?
HelixDB is an OLTP graph database with native vector and full-text search built in Rust on Object Storage.
Overview of helix-db
What it is
HelixDB is an open-source (Apache 2.0) OLTP graph database built in Rust on object storage. It provides native vector search (approximate nearest-neighbor) and BM25 full-text search. Entities, relationships, embeddings, and text are stored in a single transactional engine and can be queried together using SDKs for Rust, TypeScript, Go, or Python.
Key features
- Graph: Model entities, relationships, and typed properties as a labeled property graph.
- Vector: Approximate nearest-neighbor search, prefiltered by graph traversal.
- Full-text: BM25 keyword search over node and edge properties.
- Transactional: ACID guarantees with a single writer committing every transaction.
- Multi-language SDKs: Rust, TypeScript, Python, Go.
Use cases
- Long-term agent memory for AI applications.
- GraphRAG and hybrid RAG.
- Internal knowledge graphs that tie documents, people, tools, and decisions together.
Getting started
Install the CLI. On macOS and Linux:
curl -sSL "https://install.helix-db.com" | bash
On Windows PowerShell:
irm https://raw.githubusercontent.com/HelixDB/helix-db/main/crates/cli/install.ps1 | iex
Create a project, start a local instance (requires Docker or Podman), and run a query:
mkdir my-helix-app && cd my-helix-app
helix init local
helix start dev
helix query dev --file examples/request.json
Local data lives in memory by default; use helix start dev --disk to persist across restarts. For embedded usage without a server, see the repository README.
License and pricing
Licensed under Apache 2.0. Open source.
Based on the project's public README and repository metadata.