Google unveiled Stitch, an AI-powered tool that generates complex user interface designs and frontend code from text prompts and reference images [1].

The tool aims to reduce the manual effort required for app development by allowing creators to turn rough ideas into functional designs quickly [1]. By automating the bridge between a visual concept and the actual code, Google intends to accelerate the prototyping phase for developers [4].

Google introduced the tool during the Google I/O 2025 keynote held on May 20, 2025 [2]. The event took place at the Shoreline Amphitheatre in Mountain View, California [2]. Stitch is powered by the Gemini 2.5 Pro model, which enables the system to interpret multimodal inputs, both text and imagery, to produce structured layouts [1].

According to a Google announcement, "Stitch can turn text prompts and reference images into complex UI designs and frontend code in minutes" [6]. This capability allows developers to upload a sketch or a screenshot of an existing app and ask the AI to modify the style or add new features, which the tool then renders as production-ready frontend code [3].

The launch of Stitch is part of a broader push by Google to integrate generative AI into the developer workflow. By shifting the focus from manual coding of basic components to high-level design orchestration, the company hopes to lower the barrier to entry for app creation [5].

Stitch focuses specifically on the frontend experience, meaning it handles the visual and interactive elements that users see on their screens. This allows developers to iterate on the user experience without spending hours writing repetitive CSS or HTML for initial mockups [4].

Stitch can turn text prompts and reference images into complex UI designs and frontend code in minutes.

The introduction of Stitch signals a shift toward 'generative UI,' where the interface of an application is no longer static but can be rapidly iterated or even generated on the fly. By automating frontend code production, Google is positioning Gemini 2.5 Pro not just as a chatbot, but as a functional engineering tool that could reduce the time-to-market for new software products.