Google’s Firebase Gets AI Extensions, Opens Marketplace

Image credit: Getty Images

Firebase, Google’s backend platform-as-a-service for application developers, took a sizable share of new features at the company’s I/O developer conference today, but the highlight update was the addition of Google’s PaLM API to It is the addition of the AI ​​extension function that I used. This includes Python language support in Cloud Functions for Firebase (Gen 2), updates to the Cloud Firestore database service, and more. Additionally, Google is now opening up the Firebase extension marketplace to more developers.

Jeanine Banks, head of Google’s developer products group, said the company is focused on reducing complexity for developers. “There are so many different technologies out there, especially now that we want to build for multiple devices and build across multiple platforms, but we certainly lack the tools to help developers do that. So it’s important to make it as easy as possible to build something great,” she said. “Of course, with AI and generative AI, the risks are even higher. How can we leverage it to build great apps and experiences?”

Image credit: Google

Google first introduced Firebase extensions in 2019, a pre-packaged solution for app developers. The service is now generally available, and given Google’s current focus on AI in general, it’s no surprise that the Firebase team and the developers who use it have such concerns. Platforms — are also on board with this trend. New extensions here include an image generator, a tool to build chatbots on top of the PaLM API, a text summarizer, and more.

Until last year, Google worked directly with extension providers who wanted to offer APIs in the marketplace. Banks said it is now opening up the project to more providers, allowing them to participate.

“Our primary focus with Firebase extensions is to accelerate developer creativity and enable them to build applications,” she explained. “By enabling that, we think they’re going to succeed, their applications will succeed, and as their applications succeed, they’re more likely to scale up. We will use Play services, more payment services.” This is also how Google plans to (indirectly) monetize this service. Not through income from profit-sharing agreements, but simply by bringing more developers into the larger ecosystem of Google services.

The addition of Python as a supported language for Cloud Functions is also somewhat related to this AI effort. After all, Python is the default language for most AI projects, and as Banks points out, this allows developers in the AI/ML community to more easily build their own extensions and make them available in the market. will be

Specifically, Firebase is adding this Python support in Cloud Functions Gen 2, which is also now generally available (in public preview). Additionally, Cloud Functions 2nd Generation now offers up to 32 GB of memory, allowing you to process up to 1,000 requests in parallel.

Small updates to Firebase this year include support for OR queries in Firestore and a number of security updates, support for frameworks such as SvelteKit, Asto, and Nuxt in Firebase Hosting, support for Fultter SDK updates, and new support for Firebase Hosting. Includes support. Flutter the web using WebAssemby.