matrixorigin/matrixone

AI-native HTAP database with Git-for-Data and built-in vector search, serving as the data and memory backbone for intelligent agents and applications.

What it solves

MatrixOne is an AI-native, cloud-native HTAP (Hybrid Transactional/Analytical Processing) database designed to replace complex data stacks. It eliminates the need for maintaining separate databases for transactions (OLTP), analytics (OLAP), full-text search, and vector search, thereby removing the overhead of ETL (Extract, Transform, Load) pipelines and data synchronization issues.

How it works

MatrixOne acts as a MySQL-compatible database that integrates advanced data management features directly into its core. It uses a hyper-converged engine to handle diverse workloads in a single system. A standout feature is its "Git for Data" capability, which allows users to perform operations like instant snapshots, time travel (querying historical data), and branching/merging data, providing version control for database states. For AI applications, it includes built-in vector search (using IVF/HNSW indexing) and full-text search, allowing developers to build RAG (Retrieval-Augmented Generation) applications directly within the database.

Who it’s for

It is designed for developers and data engineers who want to simplify their infrastructure by consolidating transactional, analytical, and AI-driven workloads into a single, MySQL-compatible system. It is particularly useful for teams building AI applications that require integrated vector search and those who need robust, Git-like versioning for their data.

Highlights

  • Git-style Data Management: Supports branching, merging, instant snapshots, and time travel for data.
  • All-in-One HTAP: Unified support for OLTP, OLAP, full-text search, and vector search in one system.
  • AI-Native: Built-in vector indexing (IVF/HNSW) for RAG and semantic search without needing external vector databases.
  • MySQL Compatibility: Drop-in replacement for MySQL, enabling use with existing tools, ORMs, and applications.
  • Cloud-Native Architecture: Features storage-compute separation and Kubernetes-native deployment for elastic scaling.

Related

  • Dispatch
    LatticeDB: An Embedded Single-File Graph Database with Vector and Full-Text Search

    LatticeDB is an embedded, single-file property-graph database that integrates native vector similarity search and BM25 full-text indexing into a single query layer for local-first AI and RAG applications.

  • Project
    jeffhajewski/latticedb

    LatticeDB is an embedded, single‑file property‑graph database that also provides native HNSW vector similarity search and BM25 full‑text indexing. It offers a unified Cypher‑style query language, ACID transactions, and bindings for Python, TypeScript/Node, Go, and Java, making it suitable for local knowledge‑graph apps, agent memory, and RAG pipelines.

  • Project
    zilliztech/VectorDBBench

    VectorDBBench is a Python benchmark suite for comparing the performance, recall, and cost of many vector‑search databases and cloud services using standard insertion and search workloads.

  • Project
    voxel51/fiftyone

    FiftyOne is an open‑source Python library and web app for visual AI teams to explore, label, curate, and evaluate image/video datasets and model predictions. It provides native 2‑D/3‑D annotation, embedding visualisation, built‑in analytics (the “Brain”), and integrations with PyTorch, Hugging Face, and more. Install with `pip install fiftyone`; launch the UI via `fo.launch_app(dataset)`. The tool is aimed at computer‑vision researchers and production teams needing fast, interactive data‑centric workflows.

  • Project
    1yefuwang1/vectorlite

    Vectorlite is a Rust‑based SQLite extension that adds a virtual table for fast approximate nearest‑neighbor search on high‑dimensional vectors, using HNSW and SIMD‑accelerated distance calculations. It works on Linux, Windows and macOS Apple Silicon and is available via Python wheels and an npm package.