Google AI Studio Finally Gives Deployed Apps Clean, Shareable URLs
Google AI Studio now lets you give deployed apps a clean atelier.ai.studio URL instead of a random Cloud Run slug
- Custom URLs live: Google AI Studio apps can now get a clean
atelier.ai.studioaddress instead of a random Cloud Run slug. - Set at publish time: Choose your custom URL when clicking Publish in Build Mode, no DNS or SSL setup required.
- Free tier included: Up to 2 apps deploy for free with no credit card via the Google Cloud Starter Tier.
- Full stack backed by Cloud Run: Apps auto-scale to traffic and scale to zero when idle; API keys stay server-side and secure.
- Broader context: Part of AI Studio's ongoing push from prototype tool to full product platform, following I/O 2026 additions like Android app support and Workspace integrations.
- Try it: Build at aistudio.google.com and see a live example at atelier.ai.studio.
If you've ever deployed an app from Google AI Studio and then tried to share the link, you know the problem. The URL you get looks something like https://burning-man-animal-cuddle-614365371127.us-west1.run.app/. It's live, it works, but it's not exactly something you'd put on a landing page. Google just fixed that.
The team announced custom URLs for AI Studio deployed apps, giving them a permanent home under the atelier.ai.studio domain. Instead of a machine-generated Cloud Run slug, your app gets a clean, memorable address you choose yourself, set at publish time.
The gap this fills
Until now, the path from prompt to deployed app in AI Studio was impressively short, but the finish line was a bit rough. AI Studio lets you deploy full-stack applications directly from Build Mode, providing a fast path from prototype to a managed, scalable production environment. The infrastructure side was handled. The identity side was not.
The workaround was either living with the ugly URL or going through a multi-step process of pushing to GitHub, importing into Vercel or Cloud Run, and wiring up DNS records manually. If you've never connected a Google Cloud Run service to a third-party registrar, the "DNS Dance" can be confusing. Custom URLs on atelier.ai.studio skip all of that.
How it works
The flow is straightforward. You build your app in Build Mode, and when you hit Publish, you now get the option to set a custom URL before the app goes live. The URL lives under the atelier.ai.studio subdomain, so no DNS configuration, no external registrar, no SSL certificate to provision yourself.
The rest of the deployment stack stays the same:
- When you deploy to Cloud Run from AI Studio, your API key is included securely in the server-side environment, and the deployed app uses your API key for all users' Gemini API calls.
- Under the Starter Tier, you can deploy up to two active web applications at a time per Google Account. Cloud Run services scale automatically based on incoming traffic and scale down to zero when idle, meaning your prototypes don't consume resources when not in use.
- If your app needs user login, the Starter Tier includes Firebase Authentication with Google Sign-In preconfigured.

What it costs (and what it doesn't)
The AI Studio interface is always free with no subscription, no credits, no credit card. When you generate an API key, the Gemini Developer API also has a generous free tier. You only pay once you enable Google Cloud billing and exceed the free quota, at which point you're billed per million tokens.
The Starter Tier lets you publish up to 2 full-stack applications without setting up a Google Cloud project or billing account. If you need more than two apps or want higher quotas, you can convert your fully managed Starter Tier project into a standard Google Cloud project to unlock capabilities beyond the Starter Tier.
One thing to keep in mind: a deployed application may be publicly reachable and can generate ongoing Gemini API usage on the project owner's account. Developers still need to think carefully about abuse prevention, quotas, and access control before exposing such an application to the public internet.
The bigger picture
Custom URLs are a small feature with a disproportionate impact on how AI Studio apps are perceived. A shareable, readable link is the difference between a demo you send to a colleague and a product you put in front of users. Google has been steadily closing the gap between "prototype" and "product" in AI Studio. The platform has already added native Android vibe coding with Kotlin support, one-click deploy to Cloud Run, Firebase service integration, Google Workspace integrations, a dedicated AI Studio mobile app, and Gemini 3.5 Flash as the default model.
The custom URL feature is available now. To use it, build your app in AI Studio's Build Mode, click Publish, and set your custom URL before going live. You can see an example of a deployed app at atelier.ai.studio, where Google's own restaurant menu generator demo lives. The official deployment docs cover the full publish flow.