# 4paradigm/OpenMLDB

OpenMLDB is an open-source machine learning database that provides a feature platform computing consistent features for training and inference.

Repository: https://github.com/4paradigm/OpenMLDB
Canonical: https://ross.abutalabs.com/products/openmldb
Homepage: https://openmldb.ai
Language: C++
License: Apache-2.0
License Family: permissive
Topics: feature-engineering, machine-learning, featurestore, in-memory-database, database-for-machine-learning, machine-learning-database, feature-store, feature-extraction, database-for-ai, featureops, mlops
Last push: 2026-08-21T06:40:41+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 8, longevity 100
- inputs: {"age_days": 2000, "days_push": 12, "days_rel": 558, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1710, forks 331 (observed 2026-08-28T04:05:25.719353+00:00)

## What it is
OpenMLDB is an open-source machine learning database that acts as a feature platform, computing consistent features for both offline training and online inference using SQL. It is written in C++ and designed to reduce the data engineering effort in AI applications.

## Use cases
- compute consistent features for model training and online inference
- serve low-latency features for real-time ML predictions
- run SQL-based feature engineering pipelines
- avoid training-serving skew in feature pipelines
- backfill historical features without data leakage
- replace in-house feature platforms with an open-source alternative

## When to choose
- you need a feature store with training-inference consistency
- your team wants SQL-based feature engineering
- you need low-latency online feature serving
- you are building real-time ML applications like fraud detection or risk scoring

## When to avoid
- you only need a general-purpose OLTP or analytics database
- your features are simple enough for pandas or Spark alone
- you need a fully managed cloud feature store with vendor support
- your stack is not compatible with JVM/Python/C++ integrations

## Facets
- artifact type: service
- maturity: active
- function: database, machine-learning, etl, streaming
- domain: machine-learning, databases
- platform: self-hosted, cloud
- tags: feature-store, feature-engineering, in-memory-database, mlops, training-inference-consistency, sql-database, sql, data-engineering, linux, docker

## Member repositories
- 4paradigm/OpenMLDB (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:25.719353+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:35:02.677892+00:00, confidence not recorded.
  - readme: https://github.com/4paradigm/OpenMLDB (fetched 2026-08-28T04:05:25.719353+00:00, sha d97aad14d348)
- Data as of 2026-08-30T08:39:29.467469+00:00.