ShebaAI is a minimalist offline First-Aid assistant for Android. It bridges the gap between technology and healthcare by providing instant, AI-driven medical guidance without the need for an internet connection.
- 100% Offline Intelligence: Uses Google’s MediaPipe GenAI and Gemma-2b-it model for on-device reasoning.
- Privacy First: No data leaves your device. Everything is processed locally.
- Premium UI/UX: A clean, high-contrast Black & White interface following Material 3 guidelines.
- Smart Knowledge Base: Integrated first-aid data specifically optimized for quick symptom checking.
- Typewriter Effect: A smooth, interactive chat experience with zero-latency responses.
| Splash & Branding | Chat Interaction | Medical Guidance |
|---|---|---|
 |
 |
 |
- Language: Java (Structured for clarity and stability)
- AI Framework: MediaPipe GenAI SDK
- Model: Gemma 1.1 2B IT (Quantized)
- UI: Material Design 3 & Android XML
- Data: Structured JSON Knowledge Base
Since the AI model is large, it is not included in the repository. Follow these steps to get started:
Go to Kaggle - Gemma 1.1 2B IT and download the gemma-1.1-2b-it-cpu-int4.bin file.
- Clone the repository:
git clone https://github.com/your-username/ShebaAi.git-
Rename the downloaded model file to
model.bin. -
Place the
model.binfile inside the following directory in your project:
app/src/main/assets/
- Open the project in Android Studio Ladybug or later.
- Connect an Android device (physical device recommended, minimum 4GB RAM).
- Sync Gradle and click Run.
ShebaAI is an educational and informational tool designed to provide general first-aid guidance. It is NOT a substitute for professional medical advice, diagnosis, or treatment. In case of a life-threatening emergency, please contact your local emergency services (e.g., 999 in Bangladesh) immediately.
This project is licensed under the MIT License. You are free to use, modify, and distribute it with proper attribution.
Developed with ❤️ by Maruf Hasan