Flet 1.0 Launches: Create Web, Desktop, and Mobile Applications Entirely in Python

Flet, the open source Python framework that uses Flutter for rendering, has reached version 1.0, and the team says it is now ready for production work. The milestone comes roughly four years after the project began and matters because it lets developers write pure Python while targeting iOS, Android, Windows, macOS, Linux and the web — no Dart, Swift, Kotlin or JavaScript needed.

The release is available right now on PyPI under the Apache 2.0 license. The SDK requires Python 3.10 or newer, and installation is a single command: `pip install 'flet[all]'`.

How It Gets Tested

The team backs the 1.0 label with a broad test matrix. Framework unit tests run on Python 3.10 through 3.14, alongside tests for the Flutter layer. Control and example integration tests verify behavior and compare screenshots, catching both functional bugs and visual regressions.

Mobile gets special attention: native library tests run on Android and the iOS simulator, with the mobile pipeline building for Python 3.12, 3.13 and 3.14. `flet build` integration tests compile apps for all six platforms across those Python versions, and `flet test` launches a packaged app and drives it on all five native platforms, including Linux ARM64. Developers can mirror this setup for their own apps using pytest, `flet test` against a packaged build, and screenshot comparison on Android and iOS.

Mobile Python and Package Support

Flet bundles Python 3.12, 3.13 or 3.14 with your app, and web builds use the matching Pyodide release. The project's package index now lists more than 100 packages, including NumPy, pandas, Matplotlib, Pillow, SciPy, scikit-learn, cryptography and pydantic-core, plus the native libraries they depend on. A pipeline called mobile-forge automates wheel builds for iOS and Android, though availability still varies by package and target.

The `flet build` CLI accepts eight targets: `apk`, `aab`, `ipa`, `ios-simulator`, `windows`, `macos`, `linux` and `web`.

Performance Gains in 1.0

The release also brings measurable speed improvements. Flet now tracks changed properties and skips unnecessary comparisons during UI reconciliation, with benchmarks from the 0.83 release measuring up to 6.7x faster control diffing. In packaged native apps, a new dart-bridge lets the Python and Dart runtimes communicate inside a single process without sockets, using dedicated channels for binary data.

Packaging improved too: bytecode compilation is now on by default, and the redesigned Android packaging loads Python packages straight from the APK without extraction.

Declarative UI and AI Assistance

Flet Declarative describes the UI as a function of application state, organized into reusable components — and the team dogfoods it: Flet Studio and the Flet mobile app are themselves declarative Flet apps. The imperative style remains supported for those who prefer it.

For AI-assisted development, the Flet MCP server provides coding assistants with version-specific API information, plus tools for locating examples, icons and CLI options. Flet Studio runs in the browser with a built-in AI agent, and projects can be downloaded for local development. The project also has a documented compatibility policy: APIs are deprecated for a default of three minor releases before removal.

What to Watch

The team notes that upgrading from 0.28 is a genuine migration, since handlers now run on a single event loop. With 1.0 in hand, the question is how quickly the Python community adopts it for real cross-platform apps — and how the package index and mobile wheel coverage expand from here.

Meta description: Flet 1.0 is out: write pure Python apps with Flutter rendering for iOS, Android, desktop and web. Test matrix, mobile packages and 6.7x diffing gains.

Tags: Flet, Python, Flutter, open source, cross-platform development

Featured image: Abstract illustration of interlocking code blocks rendered in soft gradients, symbolizing one language compiling to multiple platforms — no people, logos or text.