A modern, cross-platform AI chat application built with React and Chakra UI. Oji provides an intuitive interface for AI conversations with support for both streaming and non-streaming responses.
Oji is available as native desktop applications for all major platforms:
- macOS: Native app with system integration (Intel and Apple Silicon)
- Windows: Native app with installer
- Linux: AppImage, DEB, and RPM packages
📖 For detailed installation instructions, see: Desktop Installation Guide | 桌面安装指南
Access Oji directly in your browser or deploy your own instance:
- Online Version: https://apepkuss.github.io/Oji-Assistant
- Self-Hosted: Download and deploy the web version on your own server
- PWA Support: Can be installed as a desktop app from your browser
- Mobile Friendly: Responsive design for mobile devices
📖 For detailed web installation and deployment instructions, see: Web Installation Guide | 网页版安装指南
- Node.js 18+
- npm or yarn
git clone https://github.com/apepkuss/Oji-Assistant.git
cd Oji-Assistantnpm installFor web development:
npm run devThe development server will start at http://localhost:5173
# Build for all platforms
npm run electron:dist
# Build for specific platforms
npm run build:mac # macOS
npm run build:win # Windows
npm run build:linux # Linux# Build web version
npm run build:web
# Preview web build locally
npm run preview:webThe web app will be available at http://localhost:4173/
# Build and run with Docker Compose
docker-compose up -d
# Access at http://localhost:8080- Configure your AI service endpoint in the app settings
- No default endpoint - requires manual configuration
- Supports OpenAI-compatible APIs
# Default AI service URL (for web builds)
VITE_DEFAULT_AI_SERVICE_BASE_URL=http://{host}:{port}/v1
# App version
VITE_APP_VERSION=1.0.0These environment variables can be set during build time:
# Build with custom AI service URL
VITE_DEFAULT_AI_SERVICE_BASE_URL=http://{host}:{port}/v1 npm run build:web
# Or create a .env file
echo "VITE_DEFAULT_AI_SERVICE_BASE_URL=http://{host}:{port}/v1" > .env
npm run build:webFor the web version, you can set default AI service configuration using runtime environment variables:
# Development with custom settings
AI_BASE_URL=http://your-ai-server:9068/v1 AI_API_KEY=your-api-key npm run dev
# Build with custom settings
AI_BASE_URL=http://your-ai-server:9068/v1 AI_API_KEY=your-api-key npm run build
# Preview with custom settings
AI_BASE_URL=http://your-ai-server:9068/v1 AI_API_KEY=your-api-key npm run previewAvailable environment variables:
AI_BASE_URL- Set the default AI service base URLAI_API_KEY- Set the default API key
These runtime environment variables will override any build-time defaults and provide the initial values in the Server Settings page.
For detailed usage instructions, see ENVIRONMENT_VARIABLES.md | 环境变量配置说明.
See WEB_DEPLOYMENT.md | 网页部署指南 for detailed deployment instructions including:
- GitHub Pages automatic deployment
- Docker deployment
- Static file server deployment
- PWA installation
- CORS configuration
See ELECTRON_SETUP.md | Electron 配置指南 for detailed Electron app setup and distribution.
GitHub Actions automatically builds and releases desktop apps when you create a version tag:
git tag v1.0.0
git push origin v1.0.0For production preview:
npm run preview:webThe application will be available at http://localhost:5173 (dev) or http://localhost:4173 (preview).