Skip to content

■ Getting Started

Welcome to vmbluvisual model blueprint - the graphical editor for building trustworthy, maintainable and efficient software with AI. This guide helps you install the vmblu editor (as a VS Code extension), create your first model, and run it using the built-in runtime tools.

Requirements

Before you start, make sure you have:

Installing the vmblu Extension

From the VS Code Marketplace

  • Open VS Code
  • Go to the Extensions view (Ctrl+Shift+X)
  • Search for vmblu
  • Click Install

Or install directly from the Marketplace link.

Creating Your First Model

The file extension used by vmblu for its models is .vmblu, so simply create a new file with that extension and you are ready to go.

Because .vmblu files are in json format, you can also open these files with the built-in text editor. To do that left-click on the file name in the explorer and select Open With... from the popup menu and then select the text-editor from the menu.

You can also have a look at the examples included in the distribution and play around with these.

Running a model

The model that you have built will be compiled to a runable .js file. Normally you would do this in the editor by selecting an option in the main menu, but you can also do this using a cli command. You can run that .js file using the tools that you normally use for your application, whether it is a node-based application or an application that runs in the browser. Because it is very helpfull when developping and debugging, we often use vite, but again that's just one of the possibilties. The resulting .js file can also be used as the starting point to make a bundle of course.

The format of the compiled file is given in the user-guide, but in that file the required runtime is imported and started automatically, so you do not have to do anything extra for this.

■ ⚠️ pre-1.0 Version

vmblu is currently in version 0.9.0 — nearing its first stable release.

The editor is already fully functional, and you can build, edit, and run models today. That said, as the project is being opened to a wider developer audience, new bugs and edge cases may still surface.

We welcome early adopters and feedback! Expect frequent updates as we work toward a 1.0 release within the next few months.

💬 Found something? Please open an issue or discussion on GitHub.

■ Next Steps

  • Learn the vmblu file format
  • Master the editor features
  • Run and debug with the runtime tools
  • Integrate AI using tools and MCP
  • Browse or create standard reusable nodes

■ Feedback & Community

vmblu is open source and community-driven. We welcome contributions, feature requests, and ideas:

🛠 GitHub Issues: github.com/vizualmodel/vmblu/issues

💬 Discussions: github.com/vizualmodel/vmblu/discussions

🌐 Website: https://vmblu.dev

→ vmblu