HelixDB: a database built from scratch to be the storage backend for any AI application.

Homepage | Docs | Discord | X

---

HelixDB was built on the thesis that current database infrastructure is built for how humans think about data, not AI. So we've built a database that makes it easy to build all the components needed for an AI application in a single platform.

You no longer need a separate application DB, vector DB, graph DB, or application layers to manage the multiple storage locations. All you need to build any application that uses AI, agents or RAG, is a single HelixDB cluster and HelixQL; we take care of the rest.

HelixDB primarily operates with a graph + vector data model, but it can also support support KV, documents, and relational data.

Key Features

Built-in MCP tools	Helix has built-in MCP support to allow your agents to discover data and walk the graph rather than having to generate human readable queries, letting agents actually think.
Built-in Embeddings	Don't worry about needing to embed your data before sending it to Helix, just use the Embed function to vectorize text.
Tooling for Knowledge Graphs	It is super easy to ingest your unstructured data into a knowledge graph, with our integrations for Zep-AI's Graphiti, and our own implementation of OpenAI's KG tool.
Tooling for RAG	HelixDB has a built-in vector search, keyword search, and hybrid search that can be used to power your RAG applications.
Secure by Default	HelixDB is private by default. You can only access your data through your compiled HelixQL queries.
Logical Isolation	Each Helix cluster is logically isolated in its own VPC meaning only you can ever see your data.
Ultra-Low Latency	Helix is built in Rust and uses LMDB as its storage engine to provide extremely low latencies.

Getting Started

Helix CLI

The Helix CLI tool can be used to check, compile and deploy Helix locally.

1. Install CLI

   curl -sSL "https://install.helix-db.com" | bash

2. Install Helix

   helix install

3. Setup

   helix init --path <path-to-project>

4. Write queries

   Open your newly created .hx files and start writing your schema and queries. Head over to our docs for more information about writing queries

   QUERY addUser(name: String, age: I64) =>
      user <- AddN<User({name: name, age: age})
      RETURN user
   
   QUERY getUser(user_name: String) =>
      user <- N<User::WHERE(_::{name}::EQ(user_name))
      RETURN user

5. Check your queries compile before building them into API endpoints (optional)

   # in ./<path-to-project>
   helix check

6. Deploy your queries

   # in ./<path-to-project>
   helix deploy

7. Start calling them using our TypeScript SDK or Python SDK. For example:

   import HelixDB from "helix-ts";
   
   // Create a new HelixDB client
   // The default port is 6969
   const client = new HelixDB();
   
   // Query the database
   await client.query("addUser", {
      name: "John",
      age: 20
   });
   
   // Get the created user
   const user = await client.query("getUser", {
      user_name: "John"
   });
   
   console.log(user);

Other commands:

• helix instances to see all your local instances.
• helix stop <instance-id> to stop your local instance with specified id.
• helix stop --all to stop all your local instances.

Roadmap

Our current focus areas include:

• Organizational auth to manage teams, and Helix clusters.
• Improvements to our server code to massively improve network IO performance and scalability.
• More 3rd party integrations to make it easier to build with Helix.
• Guides and educational content to help you get started with Helix.
• Binary quantisation for even better performance.

Long term projects:

• In-house SOTA knowledge graph ingestion tool for any data source.
• In-house graph-vector storage engine (to replace LMDB)
• In-house network protocol & serdes libraries (similar to protobufs/gRPC)

License

HelixDB is licensed under the The AGPL (Affero General Public License).

Commercial Support

HelixDB is available as a managed service for selected users, if you're interested in using Helix's managed service or want enterprise support, contact us for more information and deployment options.