Yes. Preswald is open source under the Apache 2.0 license and free to install and use. There's no paid tier required to build or export apps, and you can self-host the output wherever you like.

Do I need to know JavaScript to use Preswald?

No. You write apps entirely in Python using built-in helpers for text, tables and charts. The framework handles the rendering, so there's no front-end code to write.

How does Preswald run Python in the browser?

It uses Pyodide, a WebAssembly build of Python, together with DuckDB compiled to WebAssembly. Your code and queries execute in the browser tab rather than on a server, which is what lets an app run offline.

Can Preswald handle large datasets?

Only up to a point. Since processing happens in the visitor's browser, very large files or heavy computation run into memory and speed limits. For big data you'd typically pre-aggregate or sample before shipping the app.

Does Preswald have an API?

There's no documented public API for a hosted service. Preswald works through its command line tool and Python functions, and the exported app is a static bundle you deploy yourself, so most integrations happen at the code level.

How do I share a Preswald app?

Run the export command to build a static site into a dist/ folder, then share that folder or host it as a static site. The recipient opens it in a browser and everything runs locally, with no install step on their end.

Preswald

Structured Labs · Coding

Preswald is an open-source framework that turns Python scripts into interactive data apps and packages them as self-contained files that run in the browser without a server.

Visit Tool

About Preswald

What Is Preswald

Preswald is a static-site generator for interactive data apps, built by Structured Labs and released under the Apache 2.0 license. You write the app in Python, then one command builds it into a folder of static files that you can open in a browser or hand to someone else. There's nothing to install on the machine that runs it.

The problem it targets is delivery. A data scientist can build a pandas and Plotly analysis in a notebook, but sharing that work usually means asking the other person to set up a Python environment. Preswald instead bundles your code, data and UI into a portable package that runs offline. Under the hood it leans on WebAssembly, running Pyodide for Python and DuckDB for queries directly in the browser tab.

That design comes with tradeoffs worth knowing up front. Because everything runs client-side, very large datasets and heavy computation are limited by the visitor's own machine and browser memory. The project is also early, and some community-reported bugs around chart re-rendering show it's still maturing. The official site now routes to Structured Labs, which continues to develop Preswald alongside its other products.

Getting Started

Install the package with pip install preswald or uv pip install preswald.
Run preswald init my_app to generate a project scaffold with an app file, config and sample data.
Edit the Python file and import UI helpers like text, table and get_df to lay out your app.
Run preswald run to start a local dev server and preview the app at localhost.
Run preswald export to build a static bundle in a dist/ folder that runs offline and can be shared.

Product Information

A quick look at Preswald's pricing, supported platforms, and performance.

Free PlanYes

Paid PlansFree (open source)

PlatformWeb (browser), Command line

DeveloperStructured Labs

CategoryCoding

Release DateFeb 2025

Latest UpdatedOct 2025

Website VisitsN/A

Website Global Rank6.6M

API AvailabilityN/A

Best for

The users, tasks, and scenarios where this tool fits best.

Users

Data analysts
Python developers
Data teams handing work to stakeholders
Privacy-sensitive teams

Tasks

Building analyst dashboards
Writing interactive reports
Shipping data inspection tools
Packaging offline kits

Scenarios

Handing a tool to a non-technical colleague
Working with confidential data
Prototyping a quick data product
Giving AI agents a structured tool

Key features

Python-first authoring

You build the whole app in Python, using helper functions for text, tables, charts and forms. There's no JavaScript to write and no notebook runtime to fight. For people already comfortable with pandas, the move from analysis script to working app is small.

File-first export

A single preswald export command builds the app into a static site inside a dist/ folder. That output contains everything needed to run it, so you can zip it, host it or drop it into a shared drive. It works offline in any modern browser.

In-browser compute with Pyodide and DuckDB

The runtime runs Python through Pyodide and queries through DuckDB, both compiled to WebAssembly. That means real data processing happens in the tab, not on a remote server. DuckDB handles analytical SQL on local files quickly, which is what makes fast in-browser data apps possible.

Reactive engine

Preswald tracks a dependency graph of your state and only re-runs the parts of the app that a change affects. When a user moves a filter, the app recomputes just the affected outputs rather than the whole script. This keeps interactive apps responsive as they grow.

Composable UI components

The framework ships prebuilt building blocks such as tables, charts, forms and layout elements. You drop them into your script and they render into a consistent interface. That spares you from hand-writing markup and styling for common data views.

Local execution by default

Because apps run entirely in the browser, there's no backend to deploy, secure or pay for. Data stays on the machine where the app runs, which suits sensitive datasets and offline fieldwork. The same property is what limits how much data and compute an app can handle.

Open source under Apache 2.0

Preswald is licensed under Apache 2.0 and developed in the open on GitHub, so you can read the code, file issues and contribute. The project is maintained by Structured Labs with an active community on Slack and GitHub discussions.

Pros and cons

Pros

Write apps in Python, so data analysts can share their work without learning a front-end stack.
Output is a self-contained static bundle that runs offline in the browser with no server to deploy.
DuckDB and Pyodide bring real SQL and Python execution straight into the browser tab.
Reactive re-runs keep interactive dashboards responsive as they grow.
Apache 2.0 licensing means no vendor lock-in and full freedom to self-host.

Cons

Everything runs client-side, so large datasets and heavy computation hit the memory and speed limits of the visitor's machine.
The project is young, and community bug reports around chart re-rendering show the runtime is still maturing, which means rough edges are possible.
There's no documented hosted cloud API, so teams wanting a managed backend have to look elsewhere or self-host.
Browser-based delivery means apps depend on a modern browser and aren't suited to scheduled or headless server jobs.

Frequently asked questions

Preswald builds interactive data apps in Python and packages them as static files that run in the browser. It suits dashboards, reports, data inspection tools and prototypes that need to be shared without asking anyone to set up a Python environment.

Base Code

Base44 (Wix) · Coding · Productivity

Base Code is a shared cloud development environment from Base44. Teammates describe changes to an existing GitHub codebase in chat, preview them live and send them back as pull requests.

Free / $16 - $160/moView details

Google EmbeddingGemma 2

Google DeepMind · Writing · Image · Coding

Google EmbeddingGemma 2 is a 740M-parameter open model that turns text, code, images, audio, and video frames into vectors in one 768-dimension space, running on phones and laptops.

Free / Free to download under Apache 2.0, no paid tiersView details

GitGlow

GitGlow · Coding · Productivity

GitGlow is a local-first desktop Git client that lets independent developers review their own code and their AI agents' changes in one place before shipping.

Free / Free - €2.99/mo (€29/year)View details