Summary

This project provides a cross-platform Flutter application enabling users to run uncensored, 100% offline open-source AI models (GGUF format) directly on their Android, iOS, or PC devices. It targets users prioritizing total privacy, offline AI capabilities, and unrestricted model responses, offering a local alternative to cloud-based AI services without subscriptions or API keys. The primary benefit is unrestricted, private AI interaction directly on the user's hardware.

How It Works

The application leverages the Flutter framework for its cross-platform reach, integrating with llama.cpp to execute GGUF-formatted AI models locally. This approach ensures all computations and data remain on the device, guaranteeing zero cloud dependency and complete user privacy. An OpenAI-compatible local API server is also included, allowing seamless integration with external tools and clients. The design prioritizes mobile-first functionality, with desktop support actively seeking community contributions for refinement.

Quick Start & Requirements

• Android: Download the appropriate APK (e.g., app-arm64-v8a-release.apk) from the releases page and install via device settings.
• iOS: Sideload the pre-release IPA file (ios_pre_release.zip requires extraction) using TrollStore (recommended) or AltStore/AltServer. Building from source requires Xcode 15+ and the Flutter SDK.
• Desktop (Windows/macOS/Linux): Community-supported. Clone the repository, run flutter pub get, and then flutter run -d <platform>.
• Links:
  • Setup & Demo Video: https://youtu.be/2Pnv68iHIaQ
  • GitHub Repository: https://github.com/techjarves/Uncensored-Local-AI-Multiplatform.git

Highlighted Details

• Zero Censorship: Runs models designed for unrestricted responses, free from corporate safety filters or refusals.
• Total Privacy: All conversations and data are stored and processed exclusively on the local device.
• Fully Offline Operation: Models can be downloaded and used without an active internet connection.
• Local OpenAI API: Provides a compatible REST API endpoint for integration with other applications and scripts.
• In-App Model Management: Users can download, import, and manage GGUF models directly within the application interface.

Maintenance & Community

The project actively seeks community contributions, particularly for testing and polishing desktop platform support (Windows, macOS, Linux). Key roadmap items include AI Agent Mode (In Progress), Web search integration, Voice interaction, and Image/vision model support (Planned). Feedback and bug reports are encouraged via GitHub issues.

Licensing & Compatibility

The project is licensed under the MIT License, permitting free use, modification, and distribution. This license generally allows for commercial use and integration into closed-source projects without significant restrictions.

Limitations & Caveats

Desktop platform support is community-driven and may exhibit rough edges or require further testing and polish. The iOS version is pre-release and requires manual sideloading. Support for advanced features like image/vision models is planned but not yet implemented.