↑ ↓ navigate↵ selectesc close

What's new in Firebase at I/O 2025: Web Architecture & Implementation Guide

CodingCat.dev tutorial cover image
Alex Patterson
Alex Patterson

If you have ever watched an AI coding agent hallucinate deprecated Firebase v8 syntax or deploy wide-open Firestore rules, you know that raw LLM prompts are not enough for production backends. Following the announcement of What's new in Firebase at I/O 2025 by Aja Hammerly, Seba Gnagnarella on the Official Firebase Blog (2025-05-20), here is our hands-on CodingCat.dev breakdown of what actually shipped, how it works under the hood, and how we wire it into production TypeScript, Next.js, Astro, SvelteKit, React, and Angular applications. Update: As of August 2026, Vertex AI in Firebase is now called Firebase AI Logic and Vertex AI Gemini API is now called Agent Platform Gemini API.

The Mental Model: Why What's new in Firebase at I/O 2025 Matters for Web Devs

As a Google Developer Expert (GDE) for Firebase and Senior Solutions Engineer at CodingCat.dev, I evaluate every Firebase update through one lens: does it reduce boilerplate while keeping our runtime security and cloud bills predictable? Instead of relying on stale training weights, wiring up the Model Context Protocol (MCP) and structured Firebase agent skills gives your IDE or CLI direct, schema-aware tools to provision resources, inspect security rules, and run deterministic CLI deployments.

AJ Pro-Tip from Alex: Never let an autonomous coding agent push Firestore or Data Connect rules straight to production without a gate. Always require the agent to run your test suite against the Firebase Local Emulator Suite first, and audit every generated rule for strict request.auth.uid ownership checks.

How It Works: Technical Breakdown & Implementation

Google I/O is one of the most exciting events of the year, and we’re using this moment to share how we’re helping you build the next generation of apps by providing an end-to-end platform to accelerate the complete application lifecycle. Specifically, we’ve been focused on:

If you’re tuning into I/O, you can learn more about these launches in our What’s New in Firebase session, or read on for a snapshot of what’s new.

Streamline full-stack app development with Firebase Studio

We first introduced a preview of Firebase Studio, our new cloud-based AI workspace, last month at Cloud Next. With Firebase Studio, you can lean on AI assistance throughout your development to create fully functional apps with just prompts, or jump into the code thanks to the full power of a customizable IDE and underlying VM. In just over a month, the community has already created over 1.5 million workspaces!

We recently announced that Firebase Studio is now powered by Gemini 2.5, Google’s most intelligent model yet, for more feature rich apps and more polished UIs. Today, we’re unveiling new features that will give you greater control over your app’s design and help automate your backend setup.

Bring Figma designs to life in Firebase Studio Using the Builder.io plugin, you can now import your app designs from Figma into Firebase Studio, and then use Gemini in Firebase to add features and functionality. For example, Gemini can add additional screens or stitch together user flows by following your prompts in chat, without requiring you to write any code.

Image placeholder support You can now easily swap basic placeholder images for high-fidelity visuals with the App Prototyping agent. Simply click the image you want to change and then search and select a replacement from Unsplash’s extensive library.

Firebase backend integrations Another big update coming to the App Prototyping agent is the ability to detect, set up, and provision your app’s backend. If your prompt includes a database or authentication, Firebase Studio will recommend Firebase Authentication and Firestore in the App Blueprint alongside other important components like AI resources and UI elements. When you’re ready to publish your app, Firebase Studio will provision your backend services to ensure your app’s data can be stored and retrieved properly and deploy it to Firebase App Hosting.

This feature started rolling out today and all users will get access soon. If you prefer a different stack or your needs change as your app grows, you can still jump into the coding workspace and extend your apps with any backend.

To see the full list of new features for Firebase Studio and get tips on how to use them, check out this blog post.

Enhance existing apps with AI using Firebase AI Logic

At Google I/O Previously, the Firebase team launched Vertex AI in Firebase so you could harness the capabilities of the Gemini family of models directly from your mobile and web apps. We also introduced the Genkit framework, designed to give you a server-side solution for leveraging Gemini and other models. At I/O today, Firebase is bringing these investments in AI tooling, frameworks, and SDKs closer together by evolving Vertex AI in Firebase into Firebase AI Logic.

We’re also releasing a slew of new features for our client integration, this includes:

  • Direct access to the Gemini Developer API → With our client SDK, you can now use the Gemini Developer API alongside the already available Vertex AI Gemini API to directly integrate Google’s Gemini and Imagen models into your app. Plus, you can use Firebase App Check to verify client requests and Firebase Remote Config to enable dynamic updates to prompts, model parameters, and feature flags.
  • Hybrid inference with Gemini Nano on Chrome → Firebase AI Logic allows you to do on-device inference with Gemini Nano or use a cloud-hosted Gemini model if on-device capabilities are unavailable. This lets your app offer faster, more private, and potentially offline AI capabilities to users when AI tasks are executed locally.
  • Gemini API in Unity & Android XR → Firebase is releasing a preview of the Firebase AI Logic SDK for Unity so you can now integrate Gemini models directly into your games and Android XR experiences.
  • Image generation with Gemini → You could already access Imagen models client-side for text-to-image generation. Now, The Firebase team made it possible to generate and edit images using the multimodal capabilities of Gemini.
  • Expanded AI monitoring dashboard → The Firebase team added new insights into usage patterns, performance metrics, and debugging information from the Firebase AI Logic client SDKs to the AI monitoring dashboard so you can optimize your AI implementation and create better experiences for users.

While Firebase AI Logic’s client SDKs excel at simplifying client-side model access, many AI applications require server-side flexibility and integration. For these scenarios, we offer Genkit, our powerful open-source framework that supports server-side development in Firebase AI Logic. You can learn more about Genkit’s key features and recent releases here.

Optimizing development velocity with AI assistance and other updates across the platform

Transforming your app with generative AI goes beyond just adding AI-powered features - it also means accelerating your app development with AI assistance. Today, we’re giving you a first look at the Firebase Model Context Protocol (MCP) Server in the Firebase CLI.

The Firebase MCP Server allows AI assistance tools to directly interact with Firebase resources and services. With the Firebase MCP Server, you can:

  • Create Firebase projects and apps, and download the correct SDK configuration
  • Answer questions about your production app’s Firebase backend (ex. “What provider did user@example.com use to create their account?” or “List my Firestore collections, grab 3 documents from each, and use it to create a summary of my data model”)
  • Highlight the most important crashes occurring in production and their impact on users to troubleshoot issues in context
  • Trigger changes to the user environment (ex. set up databases, set auth claims, etc.)

Firebase App Hosting lets you build and deploy modern, full-stack web apps with the security and scalability of Google Cloud. Now there are two new ways to seamlessly deploy your web apps even if you aren’t using GitHub to store your source code.

  • Deploy your source code right from your local machine with the Firebase CLI. If you have other Firebase deployments like security rules in place, everything will now deploy from one command.
  • Use Terraform. Speed up rollouts and have greater control by skipping the build process entirely when you use Terraform to deploy your own container image instead of relying on App Hosting to build from your source code.

Finally, if you’re using the Firebase Emulator Suite to run your web app and backend Firebase services locally, App Hosting will now automatically use the emulated version of those services instead of production so you can safely test your app without risking your production data.

Check out more updates to Firebase App Hosting.

Firebase Data Connect gives you the ability to build secure, scalable apps with Cloud SQL for PostgreSQL. Recent updates let you:

  • Use Gemini in Firebase to simplify schema creation during onboarding and help craft and refine your queries and mutations. Work freely in the console, Firebase Studio, or VSCode with the Gemini Code Assist and Data Connect extensions.
  • Perform more sophisticated data retrieval, improve efficiency, and scale more easily with transactions that support additional server values and multi-step capabilities.

Get more information on these Data Connect launches here.

Coming soon: Firebase phone number verification

Whenever possible, we try to leave you with sneak peeks of things Firebase is working on, but not quite ready for launch. At I/O in the Chrome booth, we are showcasing a demo of phone number verification, which is coming to Firebase later this year. Soon, you’ll be able to streamline user onboarding, reduce verification costs, and eliminate SMS abuse vectors by directly obtaining verified phone numbers for your app users from their subscribed carriers! Stay tuned.

Bringing AI to every app developer

As AI continues to rapidly reshape the technology and tools you use everyday, we remain committed to helping you navigate this profound moment with confidence and ease. As AI transforms what apps are possible and how they are developed, we will continue to transform Firebase into the app development platform of the future.

You can stay up to date on the latest news and best practices on using all Firebase products by subscribing to our YouTube channel, reading our blog, and following us on X and LinkedIn.

Show Me the Code: TypeScript & Web SDK Setup

Here is the production baseline pattern we use at CodingCat.dev when wiring What's new in Firebase at I/O 2025 into a modern TypeScript web application:

# Firebase Data Connect Schema & Type-Safe Web Query
type Project @table {
  id: UUID! @default(expr: "uuidV4()")
  title: String!
  ownerUid: String! @default(expr: "auth.uid")
  createdAt: Timestamp! @default(expr: "request.time")
}

query ListUserProjects @auth(level: USER) {
  projects(where: { ownerUid: { eq_expr: "auth.uid" } }, orderBy: [{ createdAt: DESC }]) {
    id
    title
    createdAt
  }
}

Production Web Architecture & Security Checklist

Whether we are deploying on Firebase App Hosting, Cloudflare Workers, or Vercel, shipping AI features to production requires defense-in-depth across three layers:

  • Enforce Firebase App Check & Attestation: Protect your web endpoints, Cloud Functions, Data Connect GraphQL resolvers, and Firebase AI Logic calls against unauthorized replay or billing abuse using Firebase App Check with reCAPTCHA Enterprise.
  • Validate Locally with the Firebase Emulator Suite: Run deterministic integration tests against local Firestore, Auth, Functions, and Data Connect PostgreSQL emulators before promoting schema or rule changes to production.
  • Keep Server & Client Boundaries Type-Safe: Use the modular Firebase Web SDK and generated TypeScript bindings so SSR frameworks like Next.js, Astro, SvelteKit, and Angular serialize state cleanly without hydration flickers.

Conclusion

Accelerate your full-stack web and AI development by pairing What's new in Firebase at I/O 2025 with strict Security Rules, Firebase App Check, and local emulator testing. Come tell us on X (@CodingCatDev) and LinkedIn (CodingCat.dev) how you are using Firebase in your stack, and subscribe to the CodingCat.dev YouTube Channel so you don't miss what we build next!

Watch Our Related Firebase Video Tutorial on YouTube

Want to see how we architect and debug Firebase web applications live? Watch our related CodingCat.dev session Firebase SQL with Data Connect below, and make sure to Subscribe to CodingCat.dev on YouTube for weekly full-stack web and AI deep dives:

Resources, Official Docs & Related CodingCat.dev Guides

Frequently Asked Questions (FAQ)

What are the key capabilities introduced in What's new in Firebase at I/O 2025?

What's new in Firebase at I/O 2025 (announced 2025-05-20 by the Firebase team) enhances the platform for Firebase, Google Cloud, Web Dev, AI, MCP, SQL, Next.js, TypeScript, Security, Serverless workloads. Update: As of August 2026, Vertex AI in Firebase is now called Firebase AI Logic and Vertex AI Gemini API is now called Agent Platform Gemini API. Developers can review the primary announcement on the Firebase Blog alongside our implementation patterns and production checklist above.

How do web developers use What's new in Firebase at I/O 2025 with Next.js, SvelteKit, React, or Angular?

Web developers can integrate these features using the modular Firebase JavaScript/TypeScript SDK, Firebase Data Connect web bindings, and Firebase App Hosting. Always pair client-side SDK initialization with Firebase App Check and strict Security Rules, and test locally with the Firebase Local Emulator Suite before deploying.

Where can I watch video tutorials on Firebase Web and AI development?

You can watch our hands-on Firebase video walkthroughs on the CodingCat.dev YouTube Channel, including Firebase SQL with Data Connect, and follow us on X (@CodingCatDev) for weekly full-stack tutorials.