Modelizer is a lightweight database modeling tool developed for teaching the database modeling process (initially for courses in Luxembourg).
- Online version: https://www.modelizer.lu/
- Download: GitHub Releases
- Docker image: https://hub.docker.com/r/haanlaurent/modelizer
- Download the latest release from the GitHub Releases page.
- Unzip it anywhere on your computer.
- Double-click
modelizer.bat - Open your browser at http://localhost:8080
The release includes a tiny local webserver based on miniserve: https://github.com/svenstaro/miniserve
Run the latest published image from Docker Hub: https://hub.docker.com/r/haanlaurent/modelizer
docker run --rm -p 8080:80 haanlaurent/modelizer:latestThen open: http://localhost:8080
Run a specific version tag:
docker run --rm -p 8080:80 haanlaurent/modelizer:<version>Build and run locally from this repository:
docker build -t modelizer:local .
docker run --rm -p 8080:80 modelizer:localModelizer supports:
- different modeling levels:
- conceptual model (CDM)
- logical model (LDM)
- physical model (PDM)
- classes
- associations
- associative associations
- reflexive associations
- composite aggregations (enable in menu first)
- multiplicities
- attributes
- attribute types
- default values
- constraints: Null, Unique, Auto Increment
- import from Java Modelizer files
Modelizer is intentionally a teaching tool that simplifies drawing and defining models. It contains no validation logic and does not generate relationships automatically from associations. It does not help you make the "right" modeling decisions — you do.
- Export format: PNG
- Models must be downloaded manually (no server-side storage)
- Settings are stored in browser localStorage
- ChartDB (interface design inspiration): https://github.com/chartdb/chartdb/
- Java Modelizer by Bob Fisch: https://modelizer.fisch.lu/
- miniserve (local webserver used in the release bundle): https://github.com/svenstaro/miniserve