Jump to Content
Databases

AI Studio unlocks full-stack vibe coding with Cloud Run, Firebase, and Cloud SQL, no credit card required

May 21, 2026
Justin Mahood

Product Management

Gopal Ashok

Product Management

Try Gemini Enterprise Business Edition today

The front door to AI in the workplace

Try now

At Google I/O 2026, we announced  updates to the integration between Google AI Studio and Google Cloud

This is an update to the integration we announced in March, which included support for vibe-coded full-stack app deployments from AI Studio powered by Cloud Run, Firestore, and Firebase Auth

With this expanded integration, you can use AI Studio to build a broader set of applications, using either a relational database with Cloud SQL or a non-relational database with Firestore. You don’t even need to specify a database — the AI agent can infer the right database for your app or feature.

Get started today in AI Studio at no cost with Cloud Run, Cloud SQL for PostgreSQL (coming next month), Firestore, and Firebase Auth for Starter Tier.

https://storage.googleapis.com/gweb-cloudblog-publish/original_images/1-_publish.gif

Publishing a full-stack app from AI Studio to Cloud Run with a single click

An easy on-ramp: The Google Cloud Starter Tier

You can build applications in AI Studio and deploy your prototypes directly to Cloud Run, authenticate via Firebase Auth, and store your data in a Firestore or Cloud SQL database. No credit card, no Google Cloud account, no friction — just prompt and launch.

If you don’t have an account, AI Studio uses the Google Cloud Starter Tier to create resources for you. You can deploy up to two full-stack apps. If you outgrow the limits of the Starter Tier, you can upgrade to a standard Google Cloud project with a billing account. All your resources will be transferred to your billable Google Cloud project, so that your application can scale as it grows.

Powering full-stack vibe coding with Cloud SQL

We’re introducing an intelligent, automated data foundation that makes it easy for developers to focus on their applications, not their infrastructure.

AI Studio integration with Cloud SQL includes:

  • An instant on-ramp: Go from prompt to a fully-deployed PostgreSQL database rapidly with instant provisioning.

  • Zero-cost startup: Try Cloud SQL for the Google Cloud Starter Tier at no cost, without needing a credit card or Google Cloud account. 

  • Flexible cost control: The AI agent uses a new Cloud SQL for PostgreSQL developer edition, which enables the backend to scale to zero automatically, so you only pay while you’re using the app.

  • Agent-driven experience: To update your application, enter new prompts and the AI Agent automatically creates the schema and executes SQL statements in the database.

  • Global scalability: While the interface is simple, your app runs on Google Cloud’s robust, highly-reliable, and securely designed infrastructure that can scale to support millions of users.
https://storage.googleapis.com/gweb-cloudblog-publish/original_images/4_-_Cloud_SQL_AIS_Demo.gif

Creating an app powered by Cloud SQL for PostgreSQL developer edition

Full-stack vibe coding with Firestore and Firebase Auth

When you’re building an app in AI Studio, the agent proactively detects if you need data storage and authentication based on your prompt, and offers to set up a database and user authentication. For apps that benefit from a document database, the agent shows a card to turn on Firestore and Firebase Authentication with your approval.

https://storage.googleapis.com/gweb-cloudblog-publish/images/image3_BDw1RGs.max-1900x1900.png

Enable Firebase for your application when prompted by the agent

By clicking “Enable Firebase,” the agent automatically:

  • Provisions Firestore, enables authentication, and connects your app to the database

  • Creates your web app’s sign-in page and configures authentication with Google Sign In

  • Generates the Firestore code in your app so you can sync data across sessions and devices

  • Drafts and deploys Firestore Security Rules based on your app’s logic (but you should always double-check these rules before sharing or deploying your app!)

With Firebase Auth, you can:

  • Connect your apps to Google Workspace using natural language: When you ask for a feature involving Workspace (e.g. Sheets, Calendar, Gmail), the agent implements a “Sign in with Google” flow, powered by Firebase Authentication, designed to securely grant Google AI Studio access to your data.
https://storage.googleapis.com/gweb-cloudblog-publish/images/image1_MMqzmOz.max-600x600.png

Connect your app to Google Sheets, powered by Firebase Authentication

Check out more details on the What’s New from Firebase at Google I/O blog.

Getting started in AI Studio

Going from idea to app is now a reality. You can build a full-stack application at no cost using the following steps:

  1. Log into AI Studio: Access the platform to begin your project.

  2. Build with prompts: Start building your application using natural language prompts. For example, “Build an expense tracker app.”

  3. Enable the database: Prompt “Add a database” and AI Studio intelligently provisions a database through an "Enable" widget. You can explicitly ask for a relational database if you’d like to make your preference clear.

  4. Set up the system: Select “Enable” and agree to the terms.

  5. Start sharing: Deploy and share the application through the “Publish” button.

Get started today in AI Studio to turn your ideas into live applications in seconds.

Posted in