A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.
Python artificial-intelligencedagdata-sciencedata-visualizationdataflowdeveloper-toolsmachine-learningnotebookspipelinepythonreactivesqlweb-app
# /// script
# requires-python = ">=3.13"
# dependencies = [
# "marimo",
# ]
# ///
import marimo
__generated_with = "0.14.16"
app = marimo.App(width="medium")
with app.setup:
import marimo as mo
@app.cell
def decorator_wrap():
@mo.cache
def cache(x):
return x + 1
bar = cache(1)
return (bar, cache)
@app.cell
def block_wrap(mo):
with mo.cache("random") as cache_block:
x = []
a = "need a final line to trigger invalid block capture"
return (x, cache_block)
if __name__ == "__main__":
app.run()