WebPAI (Web Development Powered by AI) aims to build an AI platform for more reliable and practical automated webpage generation.
The rapid advancements in generative AI, particular multi-modal large language models, now excelling in multimodal understanding and code generation. This progress introduces a new front-end development paradigm, where multimodal LLMs can automatically convert visual designs into GUI code implementations, facilitating the software product delivery process.
This project collects a set of research resources and releases open datasets for webpage generation studies.
| Dataset Name | Description | Downloading |
|---|---|---|
| DCGen | 120 more-complicated real-world static webpages | |
| Interaction-to-code | 127 real-world user-interactive webpages and 374 interactions | |
| MRWeb | 500 real-world and synthetic multi-page resource aware webpages | |
| Design-to-code | 484 static real-world webpages | |
| DesignBench | 900 real-world webpages, three frameworks (react, vue and angular), three tasks (design generation, edit and repair) | |
| ComUIBench | 150 real-world very complex webpages with component annotations |
⭐ If you find this project useful, please give it a star.
This project is actively maintained, and we welcome your contributions :-)
Please feel free to reach out to us if you have any questions on this project.
- Yintong Huo, [email protected]
- Yuxuan Wan, [email protected]
- Jingyu Xiao, [email protected]