cypy is a CLI manga translator using YOLOv8 to detect speech bubbles and the Google Gemini API to translate foreign text, keeping artwork clean and typography well-fitted.

Before (Original Page)	After (Translated Page)

• Multi-Language Support: Translate to English, Indonesian, Japanese (with native vertical Tategaki text!), Mandarin (简体中文), Spanish, Portuguese, Javanese, and Custom Languages (supports Thai, Arabic, Cyrillic, etc. with automatic full variable font downloading!).
• Multi-Provider AI: Choose between Google Gemini, OpenAI (GPT-5.4), Zen (free, no key needed), OpenRouter (100+ models), or Custom (any OpenAI-compatible API) directly from the CLI.
• Interactive Commands: Change the target language (lang), switch API providers (provider), change models (model), or check current settings (status) on the fly inside the loop.
• Zero-Setup Startup: Prompts for the API key in the CLI and generates the .env file automatically if missing. Zen works out of the box — no API key required.
• Custom API Support: Bring your own OpenAI-compatible endpoint with configurable base URL, API key, and model.
• Auto Desktop Shortcut: Creates a rounded Windows desktop shortcut automatically on the first run.
• Persistent Settings: Your preferences (active language, provider, models, and keys) are automatically saved to data/settings.json whenever they change, preserving your configuration across runs.
• Dynamic Box Layouts: Terminal tables, guides, and status cards calculate width dynamically using East Asian Width rules, ensuring perfect column alignment even with CJK characters or long names.

Download the pre-compiled package for your OS from the Releases page.

1. Download and extract the .zip file.
2. Run the application:
   • Windows: Double-click cypy.exe inside the extracted folder. A desktop shortcut with the custom cypy icon will be automatically created on the first run!
   • Linux: Open a terminal in the extracted folder, make it executable, and run:

     chmod +x cypy
     ./cypy

   • macOS: Run ./cypy inside the extracted folder.
3. Paste your Gemini API key when prompted on the first run.

1. Clone the repo:

   git clone https://github.com/indravoyager/cypy.git
   cd cypy

2. Set up virtual environment:

   python -m venv venv
   # Activate:
   source venv/bin/activate  # Linux/macOS
   venv\Scripts\activate     # Windows (CMD)

3. Install in editable mode:

   pip install -e .

4. Run the application: Once installed, you can launch the app from any directory:

   cypy

   Or run it directly as a module from the project root:

   python -m cypy

If you want to compile cypy into a standalone, optimized directory package yourself, run the build script:

python build.py

This requires PyInstaller (which will be installed automatically if missing) and will output the ready-to-run .zip package inside the releases/ directory.

Customise settings inside the .env file:

# Gemini Config (Default)
GEMINI_API_KEY=your_gemini_api_key_here
MODEL_GEMINI=gemini-3.1-flash-lite

# OpenAI Config (Optional)
OPENAI_API_KEY=your_openai_api_key_here
MODEL_OPENAI=gpt-5.4-mini

# Zen (https://opencode.ai) — no API key required
ZEN_API_KEY=
MODEL_ZEN=minimax-m3-free

# OpenRouter Config (Optional)
OPENROUTER_API_KEY=your_openrouter_api_key_here
MODEL_OPENROUTER=qwen/qwen2.5-vl-72b-instruct:free

# Custom OpenAI-compatible provider
CUSTOM_BASE_URL=https://your-api.example.com/v1
CUSTOM_API_KEY=your_api_key_here
MODEL_CUSTOM=gpt-5.4-mini

[NOTE] Advanced layout settings (margins, font scales, etc.) can be adjusted in cypy/core/config.py.

cypy/
├── assets/              # YOLO model weights, font, and icons
├── cypy/                # Main Python package
│   ├── app.py           # Entrypoint loop & CLI logic
│   ├── __main__.py      # Module entrypoint for `python -m cypy`
│   └── core/            # Engine modules
│       ├── config.py
│       ├── translator.py
│       ├── utils.py
│       └── providers/   # LLM integrations
│           ├── base.py
│           ├── gemini.py
│           ├── openai_provider.py
│           ├── zen.py
│           ├── openrouter.py
│           └── custom.py
├── pyproject.toml       # Python package configuration
└── README.md

MIT