Infrastructure / Storage & Retrieval

OpenMLDB

OpenMLDB is a machine learning database that computes SQL-defined features consistently between offline training and online, low-latency inference. It ships as a distributed, highly-available cluster with an online and offline execution engine sharing the same feature definitions.

No PyPI artifact is declared on purpose. openmldb on PyPI is the Python SDK for connecting to a running OpenMLDB cluster, built from the `python/` directory of this repository, not the distributed database engine; installing it gives you a way to reach a cluster, not a cluster. Declaring it would make the SDK's count the product's adoption on the next warehouse refresh, so it stays undeclared and adoption reads from stars. Verified 2026-09-02 via the GitHub API and the LICENSE body.

Openness

5 high confidence

5.0

license: Apache-2.0(OSI)
source: public
core-gated: ungated

Apache-2.0 license body confirmed. The repository is public and unarchived and builds the whole cluster engine, and the README describes no paid tier, enterprise edition or license-gated build beside it.

https://github.com/4paradigm/OpenMLDB/blob/main/LICENSE recorded 2026-09-02
LICENSE file is the verbatim Apache-2.0 text.
https://api.github.com/repos/4paradigm/OpenMLDB recorded 2026-09-02
Repo metadata - license spdx_id Apache-2.0, private false, archived false, default branch main - for 4paradigm/OpenMLDB.
https://github.com/4paradigm/OpenMLDB/blob/main/README.md recorded 2026-09-02
README describes an open-source machine learning database deployed in hundreds of enterprise applications, with no paid tier, enterprise edition or license-key-gated build beside the Apache-2.0 engine.

Adoption

2 low confidence

2.0

1,711 GitHub stars, which lands in the 1K-10K stars band of the stars scale, level 2. No package artifact is declared - the PyPI openmldb package is the client SDK for a running OpenMLDB cluster rather than the cluster engine, so it is left undeclared rather than declared and ignored (see the product comments). No deployment count is published, which leaves the stars scale and its cap of 3.

https://api.github.com/repos/4paradigm/OpenMLDB recorded 2026-09-02
Repo metadata - stargazers_count = 1,711 - for 4paradigm/OpenMLDB.

Capability

4 medium confidence

4.0

Banded on the category feature matrix as complete retrieval system, distributed or cluster-capable. Placed one band below the vespa anchor - OpenMLDB is a complete distributed database with its own persistence, replication and high availability, but it retrieves and computes features rather than hosting ranking or embedding models inside a serving path, so it does not reach the top rung.

https://github.com/4paradigm/OpenMLDB/blob/main/README.md recorded 2026-09-02
README documents a distributed, highly-available machine learning database with consistent offline and online feature computation, deployed as a cluster.

Verified 2026-09-02