// the take
A Flutter plugin for rendering vector and raster maps using MapLibre (the open-source Mapbox GL fork), covering Android, iOS, and web from a single Dart API. It's for Flutter developers who need interactive maps without Mapbox's billing model or vendor lock-in. The plugin wraps maplibre-native for mobile and maplibre-gl-js for web, exposing a unified controller interface.
The three-package architecture (main plugin, web implementation, platform interface) is correct Flutter federation — you can swap the web impl or add a desktop one without touching the core. The melos-managed monorepo with code generation for layer/source properties is the right call for a spec-driven API surface that would otherwise be tedious to maintain by hand. PMTiles support is a real differentiator: it means you can ship a self-contained offline map without running a tile server. The example app is genuinely thorough — separate files per feature, not one kitchen-sink demo.
340 stars for a fork of a Mapbox plugin that's been around for years is a warning sign — adoption is thin, which means bug reports are sparse and platform edge cases go undetected. The README openly admits 'only a subset of the native SDK APIs are currently exposed', and there's no matrix showing what's missing, so you won't know what you need until you hit the wall mid-project. Offline/mbtiles support is punted to GitHub issue threads rather than documented API — that's a core use case for anyone going vendor-neutral, and shipping it as 'see issues #338 and #318' is not good enough. Desktop (macOS, Linux, Windows) is absent from the platform support table entirely, which matters for Flutter teams who actually target desktop.
We read the README, the directory tree and a key source file in June 2026. Nobody pays to be listed. How we judge →