- Local-First Architecture: Connects directly to local Ollama instances via the network. No data leaves your infrastructure.
- Privacy-Centric: Designed for security with no cloud dependency, no telemetry, and no accounts required.
- Dynamic Configuration: Full control over server endpoints and custom request headers.
- Material 3 Interface: Modern, adaptive UI supporting system dark mode and AMOLED themes.
- Advanced Model Management: Built-in selector for switching between local models (Llama, Mistral, Phi, etc.) with detailed model metadata inspection.
- Personalization: Persistent user profiles and system instructions to tailor model behavior.
- Language: Kotlin
- UI Framework: Jetpack Compose (Material 3)
- Concurrency: Kotlin Coroutines & Flow
- Dependency Injection: Koin
- Database: Room (with Kotlin Serialization for complex types)
- Navigation: Navigation 3
- Networking: Ollama-Kotlin (Ktor-based)
- Ollama Instance: Ensure Ollama is installed and running on your host machine.
- Server Configuration: Set
OLLAMA_HOST=0.0.0.0on your host to allow connections from your Android device. - Network Access: Your Android device must be on the same local network as the Ollama server.
- Download the latest build from the Releases section.
- Install on any device running Android 15 or higher.
Navigate to Settings > Connection Settings to configure your instance:
- Host URL: Specify your server's IP and port (e.g.,
http://192.168.1.100:11434). - Custom Headers: Add any necessary authentication or custom headers required by your proxy or server.
Contributions are welcome. Please adhere to standard open-source workflows:
- Fork the repository.
- Create a feature branch.
- Submit a Pull Request with a clear description of changes.
Distributed under the MIT License. See LICENSE for the full text.
Built with ❤️ by Udhay Adithya