Last month, we launched Firebase Studio, a cloud-based AI workspace, where, with a single prompt you can create a fully functional app. You can lean on AI assistance throughout, or dive into the code thanks to the full power of an underlying customizable VM that is open and extensible.
We’ve been inspired by the creativity behind the apps this community has created using Firebase Studio over the course of the last six weeks, like ShareIt, a flexible platform for sharing things in your community like tools, meals or local basketball games:
Or for some fun, Block Drop, built with just two prompts!
But you all haven’t been the only ones building! Since launch, we’ve continued to make improvements to Firebase Studio, like adding support for Gemini 2.5 Pro, updating the App Prototyping agent to show you its thinking before making changes, and fixing some of your most reported bugs
Today, we’re excited to share even more new updates – a combination of the most common requests and features to help you rapidly iterate on the next generation of AI apps.
Bring your Figma designs to life inside Firebase Studio
First up is one of the most requested features yet! We’ve partnered with Builder.io, a visual development platform, to make it easier to translate your Figma designs into functional apps. Here’s how it works.
- Use the Builder.io Figma plugin to export your designs to Firebase Studio.
- Once in Firebase Studio, use Gemini in Firebase via interactive chat to bring the app to life. For example, ask Gemini to build on the existing UI with additional screens, or add backend services like authentication to your app, without having to write any code.
Watch the video below to walk through the full experience.
Generate higher fidelity, more functional prototypes
Backend integrations
We’re introducing functionality to detect when your app needs a backend and provision it for you if your prompt includes a database or authentication. As part of the App Blueprint that details some of the most important characteristics of the app, such as its feature requirements and UI style, if needed, Firebase Studio will now recommend Firebase Auth and Cloud Firestore. When you’re ready to publish the app to Firebase App Hosting, Firebase Studio will provision those backend services for you.
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.
This feature is rolling out over the next several weeks.
Easily swap placeholder images with Unsplash!
Our Prototyper now makes it easier than ever to move from basic placeholders to high-fidelity visuals.
When you’re ready to replace a placeholder image in your app prototype, use the Select button to target the image, and you can instantly search and select a replacement from Unsplash’s extensive library. This lets you quickly create more realistic and visually appealing layouts.
Stay tuned, as we’ll be adding more image source integrations soon!
Build apps that generate images
Harness the native image generation capabilities of Gemini! Your apps can now create images based on user input. Why not try building an app that transforms user sketches into high-quality photographs?
Generate prototypes on mobile
For those moments of inspiration on the go (or an idea that strikes while you’re on the couch!), we’ve optimized the App Prototyping agent to work better on mobile devices. You can now seamlessly swap between the Preview and Chat views, making it easier than ever to build and iterate from anywhere.
Share your prompts without sharing your workspace
Use the new prototypePrompt query parameter to share your prompt directly via a URL. Try this: https://studio.firebase.google.com/?prototypePrompt=An app that creates a monthly budget
Build with the latest models and AI tools from Google
Finally, teams across Google announced major updates at Google I/O that make Firebase Studio even more powerful. Here’s how:
Build next gen apps with Gemini 2.5, Gemma, and the latest Gemini API features
Google DeepMind announced updates for Gemini 2.5 Flash and Pro models. Firebase Studio’s App Prototyping agent and interactive chat have been upgraded to leverage these latest models, allowing you to build richer, more complex applications with greater efficiency.
The Gemini API also has new features, like native audio output and Live API. Combine them with Firebase Studio’s full-stack tooling to quickly build next generation interactive and agentic applications.
In Firebase Studio’s code mode, you can also experiment with integrating the latest Gemma models into your applications for more powerful and tailored AI experiences within Firebase Studio. Watch the video below to see Firebase Studio Developer Relations Engineer, Rody Davis show how he builds apps that leverage image input with Gemma 3.
Summary
We’re rapidly building and collaborating with teams across Google and the broader ecosystem to make Firebase Studio a single place to build, deploy, and run the next generation of full-stack, agentic applications. For a full list of improvements and bug fixes available today, check out our release notes, or head over to Firebase Studio to try the latest today.
As always, keep the feedback coming, share your creations with us on social media using #FirebaseStudio, and reach out to our team in the forums with any feedback or questions. We can’t wait to see what you build with these new capabilities!