An AI-powered local study companion — built for medical students, designed for everyone.

Download · Quick Start · Providers · Screenshots

---

OpenStudy transforms your study notes into interactive visual learning experiences. It runs entirely on your machine, keeps your data private, and supports every major AI provider — from cloud APIs to local Ollama models.

---

Features

• 19 Visual Templates — Mind maps, flashcards, clinical vignettes, timelines, decision trees, concept webs, and more — generated as fully interactive HTML pages
• Multi-Provider AI — OpenAI, Anthropic, Google Gemini, Groq, OpenRouter, and local Ollama — switch providers in one click
• Up to 10 API Keys per Provider — Automatic round-robin rotation with rate-limit detection
• Document Vault — Upload PDFs, DOCX, PPTX, and images; ask questions grounded in your own files
• Gemini Context Caching — Large documents are cached at the API level, slashing token costs on repeated queries
• 6 Visual Themes — Clinical, Nightshift, Botanica, Bloom, Solstice, Arcane — visuals auto-follow the app theme
• Study Library — Curated concept cards across subjects (Medicine, CS, History, Science, and more)
• Export Chat — Save your chat sessions as PDF or PNG
• Native Desktop App — Tauri v2 (Rust + React), custom title bar, system tray, transparent window
• Privacy-first — All your notes and documents stay on your machine

---

AI Providers

Provider	Models	Get API Key
Google Gemini	gemini-2.5-flash, gemini-2.5-pro, gemini-2.0-flash, gemini-1.5-pro	aistudio.google.com/apikey
OpenAI	gpt-4o, gpt-4o-mini, gpt-4-turbo, o1, o1-mini	platform.openai.com/api-keys
Anthropic	claude-sonnet-4-6, claude-opus-4-8, claude-haiku-4-5	console.anthropic.com/settings/keys
Groq	llama-3.3-70b, llama-3.1-8b, mixtral-8x7b, gemma2-9b	console.groq.com/keys
OpenRouter	200+ models (GPT-4o, Claude, Llama, Mistral…)	openrouter.ai/keys
Ollama	Any locally running model	ollama.com — free, runs locally

Configure providers in Settings → Providers. You can add up to 10 API keys per provider and the app rotates them automatically when one hits a rate limit.

---

Tech Stack

Layer	Technology
Frontend	React 19, Vite 8, Tailwind CSS v4
Desktop Shell	Tauri v2 (Rust)
Backend	FastAPI + Uvicorn (Python 3.11+)
LLM Routing	google-genai, openai, anthropic, groq, httpx
Document Parsing	PyMuPDF, python-docx, python-pptx
Persistence	SQLite (via sqlite3)
Export	Playwright (PDF / PNG)
Visuals	19 standalone HTML templates, CSS variable theming, postMessage height/theme sync

---

Quick Start

Prerequisites

• Python 3.11 or later
• Node.js 20 or later
• Rust + Cargo — rustup.rs
• Windows: Visual Studio 2022 with "Desktop development with C++" workload

1. Clone the repo

git clone https://github.com/mohgomaa-art/NotebookMG.git
cd NotebookMG

2. Install Python dependencies

pip install -r requirements.txt

3. Install frontend dependencies

cd frontend
npm install

4. Run

# Windows — from project root:
run.bat            # Launches Tauri desktop app (full native experience)
# OR
run-browser.bat    # Browser-only mode, no Rust build required

On first launch, open Settings → Providers, pick your provider, paste your API key, and click Set as Active.

---

Installation (Pre-built)

Download the latest installer from Releases.

Run OpenStudy_x.x.x_x64-setup.exe — the bundled Python backend is included, no separate Python install needed.

---

Build Your Own Installer

# 1. Bundle the Python backend
build_installer.bat          # produces dist/OpenStudyBackend.exe

# 2. Copy the exe into the Tauri project
copy dist\OpenStudyBackend.exe frontend\src-tauri\

# 3. Build the Tauri installer
cd frontend
npm run tauri build          # produces src-tauri/target/release/bundle/

---

Project Structure

NotebookMG/
├── main.py                  # FastAPI backend entry point
├── requirements.txt
├── run.bat                  # Dev launcher (Tauri)
├── run-browser.bat          # Dev launcher (browser)
├── build_installer.bat      # PyInstaller bundler
├── frontend/
│   ├── src/
│   │   ├── App.jsx          # Main React application
│   │   └── MarkdownViewer.jsx
│   ├── src-tauri/
│   │   ├── src/lib.rs       # Rust Tauri commands + backend spawner
│   │   └── tauri.conf.json
│   └── package.json
├── services/
│   ├── llm.py               # Multi-provider LLM router
│   ├── visual_engine.py     # Visual template generator
│   ├── lean_memory.py       # SQLite chat persistence
│   ├── config.py            # Settings read/write
│   ├── text_extract.py      # PDF/DOCX/PPTX extraction
│   ├── openai_client.py     # OpenAI / Groq / OpenRouter streaming
│   ├── anthropic_client.py  # Anthropic streaming
│   └── gemini_client.py     # Gemini streaming + context caching
└── data/                    # Local user data (gitignored)
    ├── visuals/             # Generated visual HTML files
    ├── uploads/             # Uploaded documents
    └── settings.json        # User configuration

---

Screenshots

Document-grounded chat Answers cite your uploaded files	Interactive mind maps Generated from your notes
Flashcards & OSCE vignettes Memory aids and clinical practice cases	Pathophysiology flow Expandable step-by-step disease process
Mnemonic cards Visual recall aids for any topic	Disease comparison tables Structured etiology and pathophysiology breakdowns
Categorized study aids Memory tools organized by topic	Guided onboarding Walks new users through the study cycle

---

Contributing

Pull requests are welcome. For significant changes, please open an issue first to discuss the approach.

1. Fork the repository
2. Create a feature branch: git checkout -b feature/your-feature
3. Commit your changes
4. Open a pull request

---

License

MIT — free to use, modify, and distribute.

---

Built by Mohamed Gomaa

GitHub · Email · Instagram · Facebook · X / Twitter