#10 · Primary category: AI Website Builders
sketch-code
Keras model to generate HTML code from hand-drawn website mockups. Implements an image captioning architecture to drawn source images.
Project last updated:08/06/26
GitHub Stars
5.1K
Forks
677
Contributors
2
License
MIT
Why we included this project
SketchCode comes from the pix2code lineage and shows how far a single Keras model can get at converting hand-drawn wireframes into working HTML. It treats the job as image captioning: the model reads a sketch of a layout and writes the markup for it, and the repo ships pretrained weights plus 1,700 synthetic training images so you can run the conversion scripts on sample drawings right away. Training and BLEU-scoring utilities are included for anyone who wants to try their own inputs. The author is upfront that this is a proof-of-concept whose output depends on sketches resembling the bundled dataset, so it works best as a learning and prototyping reference rather than a production code generator. For anyone curious about how design-to-code models got started, this is a concrete baseline to compare today's tools against.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
wasp
A batteries-included full-stack framework for the AI era, enabling declarative development of React/Node.js web apps with built-in auth, jobs, and deployment.
puck
The visual editor for React.
kit
🧱 Describe your site, AI builds it, you own it as Markdown. Snap together Tailwind blocks like Lego — landing pages, blogs, portfolios, docs & more. No AI slop. Free to deploy anywhere 👇
webiny-js
Open-source, self-hosted CMS platform on AWS serverless (Lambda, DynamoDB, S3). TypeScript framework with multi-tenancy, lifecycle hooks, GraphQL API, and AI-assisted development via MCP server. Built for developers at large organizations.
scroll-world
A skill that turn any brand into a scrollable 3D world landing page