Open-source embedding database for AI apps, storing documents and vectors to power semantic, hybrid and full-text search.
- Stars
- 6,220
- Forks
- 391
- Open issues
- 25
- Last commit
- yesterday
What is it?
HelixDB is an Apache 2.0 licensed graph database that keeps entities, relationships, embeddings and text in one transactional engine. It models data as a labeled property graph, runs approximate nearest-neighbor vector search that can be prefiltered by graph traversal, and provides BM25 keyword search over node and edge properties.
The database runs locally through a CLI, either as a server on port 6969 (in memory by default, with a disk option) or embedded inside your own process. Queries are written with SDKs for Rust, TypeScript, Python and Go and sent as JSON to the POST /v2/query endpoint. Cypher reads and CRUD are supported in source builds.
Helix Cloud offers a managed deployment where object storage serves as the durable store, a single writer commits transactions, and reader nodes scale with query load.
Key features
- 01Labeled property graph for entities, relationships and typed properties.
- 02Approximate nearest-neighbor vector search prefiltered by graph traversal.
- 03BM25 full-text search over node and edge properties.
- 04SDKs for Rust, TypeScript, Python and Go sending queries over HTTP.
- 05Local server via CLI, with optional disk persistence or embedded mode.
- 06Cypher reads and CRUD in source builds.
Use cases
- Retrieval-augmented generation that combines vector similarity with graph relationships.
- Applications needing graph queries and keyword search over the same data.
- Prototyping a data model locally with an embedded or in-memory database.
- Teams wanting a managed graph database deployed on object storage.
Quick start
- 1
Install the CLI on macOS or Linux
- 2
Create a project directory and initialize a local project
mkdir my-helix-app && cd my-helix-app - 3
Start a local instance (requires Docker or Podman)
helix start dev - 4
Run the generated example query
helix query dev --file examples/request.json
Things to know
- Requires Docker or Podman to run a local instance.
- Current SDK and HelixDB generation is v3, and the wire endpoint is still /v2/, so APIs may still be evolving.
- Cypher support is documented for source builds only.
Topics
FAQ
- What is helix-db?
- An open-source graph database in Rust that stores graph, vector and full-text data together on object storage, queryable from Rust, TypeScript, Go and Python. HelixDB is an Apache 2.0 licensed graph database that keeps entities, relationships, embeddings and text in one transactional engine. It models data as a labeled property graph, runs approximate nearest-neighbor vector search that can be prefiltered by graph traversal, and provides BM25 keyword search over node and edge properties.
- Is helix-db free to use? What license does it have?
- helix-db is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is helix-db still maintained?
- Actively maintained. The last commit to helix-db was on Oct 20, 2026.
- What can I use helix-db for?
- Retrieval-augmented generation that combines vector similarity with graph relationships. Applications needing graph queries and keyword search over the same data. Prototyping a data model locally with an embedded or in-memory database. Teams wanting a managed graph database deployed on object storage.
- How do I install helix-db?
- Install the CLI on macOS or Linux Create a project directory and initialize a local project: mkdir my-helix-app && cd my-helix-app Start a local instance (requires Docker or Podman): helix start dev Run the generated example query: helix query dev --file examples/request.json
- Where can I get helix-db?
- The source code and installation instructions are on GitHub: https://github.com/HelixDB/helix-db
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
An AI-native search database that stores vector, full-text, structured and semi-structured data in one MySQL-compatible engine, for agent memory and retrieval.
Open-source AI memory platform that turns documents, code and conversations into a self-hosted knowledge graph for AI agents.
A Rust library for building LLM applications, plus a terminal coding agent built on it whose features are Bevy plugins.
An embedded retrieval library and database for multimodal AI, offering vector, full-text and SQL search over data stored in the Lance format.
A graph database that stores adjacency as sparse matrices and queries with OpenCypher, aimed at knowledge graphs for LLM and GraphRAG workloads.