Next AI Draw.io lets you create and edit draw.io diagrams simply by chatting with an AI. Instead of dragging shapes onto a canvas by hand, you describe the system, flowchart, or architecture you want, and the AI generates and refines the diagram for you in valid draw.io format.
Next AI Draw.io is an open-source project on GitHub with around 36,000 stars, actively maintained with updates in 2026. It is built with Next.js and ships in three forms: a hosted web app, a cross-platform desktop app built with Electron, and an MCP server that brings diagram generation into AI coding tools. You connect the large language model of your choice, describe what you need, and iterate on the diagram through conversation — the AI can create new diagrams, edit existing ones, and reorganize layouts while keeping the underlying XML editable in draw.io itself.
Because the output stays in draw.io’s native format, nothing is locked in: you can open any AI-produced diagram in the regular draw.io editor and keep polishing it by hand. The project can be deployed on Vercel or with Docker for teams that want their own instance, and it keeps a history of AI edits so you can compare versions and roll back a change that went in the wrong direction.
Next AI Draw.io is free and open source under the Apache-2.0 license. You can use the hosted web version or self-host it; the only cost is your own LLM API usage if you connect a paid model provider.
Software architects, developers, and technical writers who produce architecture diagrams, flowcharts, and system documentation — and would rather describe a diagram in words than draw it box by box.