Welcome to MasterG 📚
MasterG is a lightweight, multilingual AI-powered educational platform that generates curriculum-aligned content in 22+ Indian languages. Designed for 100% offline operation with zero reliance on cloud services, MasterG runs efficiently on low-resource devices (4-8GB RAM) without requiring internet connectivity.
🏆 MasterG Won First Place at Eduthon Techfest IIT B 29th Edition 🎉
- React 19 with TypeScript
- Vite
- Tailwind CSS
- React Router DOM
- Recharts (Data Visualization)
- KaTeX (Math Rendering)
- HTML5 Canvas API
- Node.js + Express with TypeScript
- MongoDB (Local session management)
- ChromaDB (Local vector embeddings)
- Tesseract.js (OCR)
- Ollama (Local AI inference)
- Expo + React Native
- Expo Router
- Expo Camera
- Expo SQLite (Local storage)
- llama.rn (On-device AI processing)
- DeepSeek-R1 1.5B (Local reasoning model)
- NLLB-200 600M (Local multilingual translation)
- ChromaDB (Local vector database)
- Tesseract.js (Local OCR)
- AI Whiteboard (AI-Board) 📝 - Canvas-based drawing with AI-powered card operations
- AI Chat Interface 💬 - Document upload with RAG-powered multilingual chat
- Learning Material Resource (LMR) 📚 - Generate summaries, quizzes, and revision notes
- Educational Poster Generation 🖼️ - Culturally relevant educational posters
- Offline Content Generator (Stitch) 🌎 - Real-time multilingual content generation
- Mobile Application 📱 - Cross-platform offline-capable mobile app
- Clone this repository:
git clone https://github.com/ayuxy027/MasterG.git
- Install frontend dependencies:
cd frontend npm install - Install backend dependencies:
cd ../backend npm install - Setup Ollama (Required for AI features):
# Install Ollama from https://ollama.ai # Pull required models ollama pull deepseek-r1:1.5b
- Setup ChromaDB (Required for vector storage):
# ChromaDB runs locally, no cloud setup needed # Follow ChromaDB installation guide
- Start backend server:
cd backend npm run dev - Start frontend development server:
cd frontend npm run dev
Note: MasterG is 100% offline and requires no external API keys or cloud services. All AI processing happens locally using Ollama and local models.
Contributions are welcomed! 😁
Let's keep the spirit of open source alive!
- Please follow the project's license terms before redistributing or modifying.
- Creating something from scratch takes a lot of hard work, and I hope you respect that. 😄
- If you have any questions or need permission for something specific, feel free to reach out.
If you have any inquiries or suggestions, please reach out:
- Email: [email protected]
- X (Twitter): ayuxy027
- GitHub: ayuxy027