Google Labs has officially upgraded its experimental design tool Stitch to a fully functional AI software design platform. The tool allows users to generate user interfaces directly through natural language instructions, and Google refers to its core logic as "Vibe Design."

image.png

The emergence of Stitch has changed the starting process of software development, making it accessible for non-professionals to get started quickly:

  • Unlimited Canvas Interaction: The platform provides an infinitely extendable canvas, allowing users to drag images, text, and code as context directly into it, and the AI will understand the design intent based on this.

  • Multiple Path Exploration: The new built-in design agent (Design Agent) can analyze the entire project and generate multiple different design options for the user to choose from simultaneously.

  • Real-Time Voice Adjustments: It supports real-time modifications to the canvas content through voice commands, greatly enhancing the intuitiveness of the interaction.

  • One-Click Prototype Conversion: Static design drafts can be directly converted into clickable interactive prototypes, and support sharing design rules between different tools using the new DESIGN.md format.

Google has targeted professionals and entrepreneurs without any design background with Stitch: users don't need to start from scratch drawing wireframes; they just need to describe the "feeling" and "experience" they want their product to convey. Developers can integrate it into professional tools like AI Studio via the MCP server and SDK, bridging the final gap from design to development.

Stitch is now available in all regions where Gemini is available. Users aged 18 or older can access its official site to try it out.

Through this platform, Google is trying to transform UI design from a tedious pixel-level task into a creative expression based on natural language.