All Tools
M
DataFreeOpen Source
MARIMO
Reactive Python notebook for reproducible data science and interactive apps
Apache-2.0
ABOUT
Traditional Jupyter notebooks have fundamental flaws: cells can be run in any order producing unreproducible state, outputs are hidden behind binary .ipynb files that break git diffs, and there is no clean path from exploration to deployment. Marimo solves these by making notebooks reactive — cells automatically re-run when their inputs change, eliminating hidden state. Output is pure Python (no binary format), enabling proper version control. Notebooks can be run as scripts or deployed as interactive web apps without rewriting.
INSTALL
pip install marimoINTEGRATION GUIDE
1. Build reactive data pipelines where changing an input cell automatically propagates through all downstream computations
2. Create interactive dashboards and data apps from notebook cells without switching to a separate framework
3. Version control data science work properly with clean git diffs on pure Python notebook files
4. Write SQL queries against databases and dataframes alongside Python transformations in the same notebook
TAGS
pythonnotebookreactivedata-sciencevisualizationsqlreproducible