Repolode
Submit
dingodb

dingodb/dingo

An open-source distributed vector database that combines MySQL-compatible SQL with vector search over structured and unstructured data.

Open-source alternative to:Pinecone

View on GitHubWebsite
Stars
1,704
Forks
263
Open issues
8
Last commit
29 days ago

What is it?

DingoDB is a distributed multi-modal vector database developed by DataCanvas. It puts relational tables and vectors in one system, so applications can store structured records and embeddings together and run upserts and vector queries through a unified SQL interface that is compatible with the MySQL protocol.

The database supports hybrid retrieval that combines traditional scalar indexes with vector indexes, and it builds indexes in real time. Data is split into shards automatically, with shards split and merged as data grows or shrinks. High availability is built in, so no external components are needed for replication. For very large datasets, it offers disk-based vector search to reduce memory use and can switch between index types depending on data scale.

Access is available through SQL, SDKs and APIs. The project provides Docker and Ansible deployment guides, and its storage layer is handled by the companion Dingo-Store project, which uses the Raft protocol for strong consistency.

Key features

  • 01Unified SQL interface compatible with the MySQL protocol for structured and vector data.
  • 02Hybrid retrieval combining scalar indexes and multiple vector index types.
  • 03Automatic data sharding with dynamic shard size, splitting and merging.
  • 04Built-in high availability without external components.
  • 05Real-time index building in the background.
  • 06Disk-based vector search for large datasets with lower memory use.

Use cases

  • Running semantic search where embeddings and business records live in the same database.
  • Building retrieval-augmented applications that filter vectors by structured fields.
  • Storing large embedding collections that exceed available memory.
  • Replacing a separate relational database and vector store with one system.

Quick start

  1. 1

    Read the official documentation for setup options

  2. 2

    Deploy on a single node with Docker, following the linked guide

Things to know

  • The README gives no direct install command, so setup depends on external docs.
  • Deployment is documented mainly for Docker or Ansible, which means some infrastructure work is needed.

In collections

Databases & SQL tools

Topics

#vector-database#embedding-search#hybrid-search#mysql-compatibility#distributed-database#semantic-search#sql#embedding-store#key-value-distributed-store#real-time-semantic-search

FAQ

What is dingo?
An open-source distributed vector database that combines MySQL-compatible SQL with vector search over structured and unstructured data. DingoDB is a distributed multi-modal vector database developed by DataCanvas. It puts relational tables and vectors in one system, so applications can store structured records and embeddings together and run upserts and vector queries through a unified SQL interface that is compatible with the MySQL protocol.
Is dingo free to use? What license does it have?
dingo is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is dingo still maintained?
Actively maintained. The last commit to dingo was on Sep 23, 2026.
What can I use dingo for?
Running semantic search where embeddings and business records live in the same database. Building retrieval-augmented applications that filter vectors by structured fields. Storing large embedding collections that exceed available memory. Replacing a separate relational database and vector store with one system.
How do I install dingo?
Read the official documentation for setup options Deploy on a single node with Docker, following the linked guide
Where can I get dingo?
The source code and installation instructions are on GitHub: https://github.com/dingodb/dingo

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects