4paradigm/OpenMLDB
An open-source feature store and ML database that computes consistent features for both model training and online inference.

OpenMLDB is a machine learning database purpose-built to solve feature consistency challenges between offline training and online inference. It provides a feature platform that enables feature engineering, feature extraction, and feature serving with consistent results across batch and real-time workloads. The system supports both batch and streaming data sources and is designed to be used as part of MLOps pipelines for feature operations.
Frequently asked
- What is 4paradigm/OpenMLDB?
- An open-source feature store and ML database that computes consistent features for both model training and online inference.
- Is OpenMLDB open source?
- Yes — 4paradigm/OpenMLDB is open source, released under the Apache-2.0 license.
- What language is OpenMLDB written in?
- 4paradigm/OpenMLDB is primarily written in C++.
- How popular is OpenMLDB?
- 4paradigm/OpenMLDB has 1.7k stars on GitHub.
- Where can I find OpenMLDB?
- 4paradigm/OpenMLDB is on GitHub at https://github.com/4paradigm/OpenMLDB.