This project is dedicated to exploring high-performance large language model capabilities on mobile devices, based on the llama.cpp project.

• Run LLMs directly on Android devices
• Native C++ implementation using Android NDK
• Context compression for long text processing:
  • Support longer text with smaller memory footprint
  • Efficient context management for extended conversations
• Simple Android UI for interaction

• Model list management (including model descriptions, download links, etc.) is currently hardcoded in the source code. This will be improved in future updates with a more flexible and maintainable solution.
• Future improvements planned:
  • Better model management system
  • Performance optimizations for mobile scenarios

• Android Studio Arctic Fox (2020.3.1) or newer
• CMake 3.18.1 or newer
• Android NDK 21.4.7075529 or newer
• Android SDK Platform 21 or newer

1. Clone the repository:

git clone https://github.com/linkaisheng/llama.cpp.android.git
cd llama.cpp.android

2. Open the project in Android Studio

3. Sync project with Gradle files

4. Build the project using Android Studio or run:

./gradlew assembleDebug

This project is licensed under the same terms as llama.cpp - MIT License. See the LICENSE file for details.

• llama.cpp - Original llama.cpp project
• All contributors to the llama.cpp project

This is an unofficial port of llama.cpp to Android. Please ensure you comply with the model's license terms and usage restrictions.