This project demonstrates Natural Language UI Programming (NLUP) and Large Language Model Rendering (LLMR) using GPT-3.5, enabling users to generate and edit web interfaces via natural language prompts. It targets developers and power users interested in the future of AI-driven software development, offering a glimpse into a paradigm where LLMs directly create and manage user interfaces and APIs.

How It Works

The core approach leverages GPT-3.5 to translate natural language descriptions into front-end UI code. Users input a description, and the model generates the corresponding HTML, CSS, and JavaScript. The system allows for iterative refinement by enabling users to select UI components and provide further natural language instructions for modification, all handled client-side. This front-end-centric rendering model aims to streamline UI development and customization.

Quick Start & Requirements

• Install and Run: npm install followed by npm run dev
• Prerequisites: Node.js environment.
• Demo Video: https://www.bilibili.com/video/BV19X4y1r762, https://www.youtube.com/watch?v=gG1I5SnrdNc

Highlighted Details

• Purely front-end implementation.
• Utilizes GPT-3.5 for code generation.
• Supports iterative UI editing via natural language.
• Explores the concept of Large Language Model Rendering (LLMR).

Maintenance & Community

The project is a personal exploration by lxfater. Community interaction and optimization discussions are encouraged via GitHub Issues. Further insights can be found on the author's Twitter: https://twitter.com/lxfater.

Licensing & Compatibility

The repository does not explicitly state a license. Users should exercise caution regarding commercial use or integration into closed-source projects until a license is clarified.

Limitations & Caveats

The author explicitly warns that dynamic UI code is generated by an OpenAI model and cannot guarantee safety from XSS attacks. Users are strongly advised to use private browsing mode and take necessary protective measures, especially if they do not fully understand the implications of running AI-generated code. The project is presented as an immature idea and exploration with significant optimization potential.