LocalSite AI is a modern web application designed to generate HTML, CSS, and JavaScript code for web pages and components using natural language prompts. It targets developers and designers seeking a rapid prototyping tool that leverages AI, offering both local and cloud-based AI model integration for flexibility. The primary benefit is the ability to quickly translate textual descriptions into functional web UIs.

How It Works

The application utilizes a Next.js 15 frontend with React 19 and Tailwind CSS for a responsive, dark-themed UI. It integrates with AI models via the OpenAI SDK, supporting local execution through Ollama and LM Studio, or cloud providers like DeepSeek and any OpenAI-compatible API. Users input a prompt, select an AI provider and model, and the AI generates the code, which is then displayed in a live preview with adjustable viewports and an integrated Monaco code editor for direct modification.

Quick Start & Requirements

• Install: Clone the repository, run npm install (or yarn install).
• Prerequisites: Node.js (v18.17+), npm/yarn, Ollama/LM Studio OR an API key for supported providers.
• Configuration: Rename .env.example to .env.local and configure API keys/endpoints.
• Run: npm run dev (or yarn dev).
• Demo: Watch Demo Video

Highlighted Details

• Supports local AI models via Ollama and LM Studio.
• Integrates with DeepSeek and any OpenAI-compatible API.
• Features a live preview with desktop, tablet, and mobile views.
• Includes an integrated Monaco code editor for direct code editing.
• Supports "Thinking Models" for advanced code generation.

Maintenance & Community

The project is actively maintained by weise25. Contributions are welcome via pull requests. Further community engagement channels are not explicitly listed in the README.

Licensing & Compatibility

The project is licensed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

Local models cannot be used directly when deployed on cloud platforms like Vercel or Netlify without tunneling solutions (e.g., ngrok). The roadmap indicates planned support for more AI providers and advanced features like file-based generation and a desktop app.