Galileo AI homepage

Galileo AI

by Google (formerly Galileo AI, Inc.)

Text-to-UI generator that turned prompts into editable mockups, now Google Stitch

Product with AI agentsAssistantdeprecated

Last reviewed 2026-06-20

Galileo AI was a text-to-UI design tool that turned plain-language prompts (and sketches, screenshots, or wireframes) into high-fidelity, editable mobile and web interface mockups, then let designers export them to Figma. Founded in late 2022 by Arnaud Benard and Helen Zhou, it ran a private beta in October 2023 and a public beta from February 2024, positioning itself as a way to get from idea to a polished first draft of a UI in under a minute. Google acquired Galileo AI in 2025 and folded its technology into Google Stitch, a Gemini-powered UI design tool launched in Google Labs at Google I/O on May 20, 2025. The standalone Galileo AI product was wound down: the original usegalileo.ai domain now permanently redirects to stitch.withgoogle.com, and Galileo users were given a migration window to move their work to Stitch. This entry documents Galileo AI as a deprecated product; its living successor is Stitch.

What it can do

  • Generate UI mockups from a text prompt

    Assistant

    Turned a plain-language description (for example, a dashboard for dog walkers) into multiple high-fidelity, editable mobile and web UI layouts, reportedly within about a minute.

    source
  • Convert sketches and screenshots into UI (image-to-UI)

    Assistant

    Took sketches, screenshots, or wireframes as visual input and produced polished interface designs, a feature added after the private beta.

    source
  • Export generated designs to Figma

    Assistant

    Sent AI-generated designs to Figma so designers could refine and ship them in their existing workflow.

    source
  • Continued as Google Stitch (Gemini-powered UI generation and code export)

    Assistant

    The technology lives on in Google Stitch, which generates responsive UIs from prompts, sketches, or screenshots and exports front-end code (HTML/CSS, React) as well as Figma. Stitch reportedly uses Gemini models (Gemini 2.5 Flash for standard output, Gemini 2.5 Pro for higher-fidelity output).

    source

Strengths

  • +Pioneered fast, prompt-to-high-fidelity UI generation with one-step Figma export
  • +Image-to-UI let designers turn sketches and screenshots into editable mockups
  • +Technology and team continue inside Google Stitch, which is free in Google Labs and adds code export

Limitations

  • Deprecated as a standalone product: usegalileo.ai now redirects to Google Stitch
  • Output was a first-draft design that a human still had to edit, arrange, and ship (assistant, not an agent)
  • Users had to migrate their work to Stitch; the original Galileo accounts and pricing no longer apply

Overview

Galileo AI was a text-to-UI design tool: describe an interface in plain language, or feed it a sketch or screenshot, and it generated high-fidelity, editable mobile and web mockups you could export to Figma. Founded in late 2022 by Arnaud Benard (ex-Google) and Helen Zhou (ex-Facebook, ex-Cruise), it ran a private beta in October 2023 and opened a public beta in February 2024.

In 2025 Google acquired Galileo AI and absorbed its technology into Google Stitch, a Gemini-powered UI design tool launched in Google Labs at Google I/O on May 20, 2025. Galileo AI no longer operates on its own: the original usegalileo.ai domain now permanently redirects to stitch.withgoogle.com. This entry treats Galileo AI as a deprecated product and points to Stitch as its successor.

What it does

Galileo AI generated multiple UI layout options from a single prompt, reportedly in about a minute. It supported image-to-UI, turning sketches, screenshots, or wireframes into polished designs, and exported generated designs to Figma for refinement. Across all of these, the AI produced a draft and the designer edited and shipped it, so it functioned as an assistant rather than a multi-step agent. The successor, Google Stitch, keeps the prompt-to-UI and image-to-UI flow and adds front-end code export (HTML/CSS, React) on top of Figma export, reportedly running on Gemini 2.5 Flash and Gemini 2.5 Pro.

Integrations & setup

Galileo AI's main integration was Figma export. It was self-serve SaaS used in the browser. After the acquisition, usage moved to Google Stitch, which is accessed in Google Labs with a standard Google account.

Pricing

Galileo AI ran a freemium beta during its independent life; its standalone pricing no longer applies after the Google acquisition. Its successor, Google Stitch, has been offered free in Google Labs (as of mid-2026), with paid tiers reported as planned but not the subject of this entry.

Best for / not for

Historically a fit for designers and product teams who wanted a fast first draft of a UI from a prompt or sketch, with a clean handoff to Figma. It was not a fit for anyone wanting hands-off, end-to-end autonomous design, or for teams that needed a stable standalone vendor (it has since been folded into Google Stitch).

Alternatives

Uizard offers prompt-to-UI and screenshot-to-design aimed at non-designers; v0 and Lovable generate working app UIs and code from prompts; Relume builds AI site structure and components for designers. Google Stitch is the direct successor.

What people are saying

We aggregate real LinkedIn discussion into sentiment for the agents people search most. Galileo AI isn't tracked yet, want it added? Request tracking.

FAQ

Is Galileo AI still available?+

Not as a standalone product. Google acquired Galileo AI in 2025 and rolled it into Google Stitch, a Gemini-powered UI design tool in Google Labs. The original usegalileo.ai domain now permanently redirects to stitch.withgoogle.com, and Galileo users were asked to migrate their work to Stitch.

Was Galileo AI an autonomous design agent?+

No. It was a generative text-to-UI assistant: it produced editable UI mockups from a prompt or image, but a designer still edited, arranged, and shipped them. It sat at the assistant level, not a multi-step agent.

Who founded Galileo AI and how much did it raise?+

It was founded in late 2022 by Arnaud Benard (ex-Google) and Helen Zhou (ex-Facebook, ex-Cruise). It raised a $4.4M seed round led by Khosla Ventures (reported February 2024) before being acquired by Google.

Sources

Last reviewed 2026-06-20

Alternatives & related