This repository provides a Dart FFI binding for llama.cpp, targeting Flutter mobile apps on iOS, Android, and macOS. It wraps the native C API with a modular, isolate-safe architecture: a worker isolate (LlamaEngine) owns the native model and context, while the main isolate communicates via message passing. The binding exposes streaming token generation, chat template rendering, multimodal (vision/audio) support, speculative decoding, context shifting, and session persistence. The codebase is well-structured with clear separation of concerns: FFI bindings, model/context wrappers, sampling, tokenization, and generation logic are each in their own modules. It uses modern Dart features (sealed classes, records, pattern matching) and demonstrates high craftsmanship in error handling, resource management, and cross-isolate communication. The package is a Flutter plugin with native-assets build hooks for Android and Swift Package Manager support for Apple platforms.
Fonnx is a Flutter plugin that enables cross-platform, on-device machine learning inference by wrapping the ONNX Runtime engine. It provides a unified Dart API that abstracts away the complexities of native FFI (for iOS, Android, macOS, Windows, Linux) and WebAssembly (for web), allowing developers to run models like OpenAI Whisper, Silero VAD, Magika, and various text embedding models without cloud dependencies. The codebase exhibits high software craftsmanship. Its architecture uses conditional Dart imports to cleanly separate native and web backends, employs Dart isolates for offloading heavy inference tasks, and implements a custom RPC protocol for web workers. A standout feature is its robust native memory management, including a C finalizer tested with sanitizers to prevent leaks during Flutter hot-restarts. While primarily a software library, it bridges the gap between ML research and production by bundling specific, optimized ONNX models and providing a deterministic test corpus with golden outputs. It is an excellent resource for Flutter developers building privacy-focused, offline-first AI applications, though its reliance on pinned native assets and specific build tooling introduces a moderate technical drift risk.
This monorepo provides a Flutter plugin ecosystem for creating native home screen widgets on iOS (WidgetKit) and Android (Jetpack Glance). It consists of three core packages: `home_widget` (Dart API for data persistence, updates, scheduling, and background interactions), `home_widget_generator` (annotation-based code generation emitting SwiftUI and Kotlin code from `@HomeWidget` annotations), and `home_widget_cli` (command-line tool for scaffolding and patching native projects). The plugin handles cross-platform widget lifecycle, configuration, localized text, timed data, interactive clicks, image assets, and widget previews. Multiple example apps demonstrate features like configurable widgets, file sharing, lock screen widgets, and generator-based layouts. The architecture is a developer toolchain with a monorepo structure using Melos, extensive CI/CD workflows, and comprehensive unit and integration tests.
This repository is a Flutter plugin that bridges Flutter applications with Apple CarPlay and Android Auto in-car infotainment systems. It provides a Dart API to construct native vehicle-optimized UI templates (lists, grids, tab bars, alerts, search, etc.) and manage navigation stacks on both platforms. The architecture is a classic Flutter plugin pattern: Dart models serialize to JSON, which is passed via method channels to native iOS (Swift) and Android (Kotlin) implementations that render the corresponding native CarPlay/Android Auto templates. Event channels handle user interactions (list item selection, alert actions, connection changes) back to Dart. The plugin includes a sophisticated SVG rasterization utility that converts Flutter asset SVGs to PNG bytes for native consumption, since native platforms cannot render SVG directly. It also provides image loading with caching and tinting on both platforms. The example app demonstrates full setup and usage for both iOS and Android, including scene delegate configuration and Android Auto service registration. External dependencies include the Flutter SDK, flutter_svg for rasterization, and native frameworks (CarPlay.framework on iOS, Car App API on Android).
This is a Flutter package for creating step-by-step showcase/onboarding tours. It highlights widgets with tooltips, supports custom tooltip widgets, auto-play, animations, and multiple scopes. The architecture uses a service locator pattern (ShowcaseService) to manage independent showcase scopes without requiring BuildContext. The overlay system uses a custom ShapeClipper to cut out holes around target widgets. The code is well-organized and actively maintained, though it lacks unit tests.
This is the 'Awesome Flutter' curated list repository — a community-driven directory of Flutter libraries, tools, apps, and learning resources organized into a single large markdown file (source.md / README.md). It is not a software application, library, or data pipeline. It contains no executable source code, no API endpoints, no tests, no CI/CD, and no dependency manifests. The entire repository is a markdown document with contribution guidelines, a PR template, a funding config, and a single image asset. The repository's sole function is to serve as a human-curated index of external Flutter resources. The 'code' is markdown links organized by category (UI, navigation, state management, plugins, open-source apps, etc.). There is no architecture, no data flow, no entry point, and no logic to analyze. The README.md is marked as linguist-generated and is effectively a rendered copy of source.md. Maintenance appears stale: the last commit is recent but the content is a static list that requires manual curation. There is no automated validation of links, no testing framework, and no build system. The contribution process is entirely manual via pull request. The repository serves a discoverability purpose for the Flutter ecosystem but contains zero reusable software components.