Google AI Studio

About

Google AI Studio is Google's browser workspace for building with the Gemini API. Its page title is "Build with the Gemini API" and its tagline is "The fastest path from prompt to production with Gemini". Build mode, documented on Google's developer site, lets you create web apps or native Android apps from a natural-language prompt, with a live preview. Google's Gemini API pricing page lists Google AI Studio access as part of the free tier, and as of October 2026 some Gemini models are free of charge while others, such as Gemini 4 Argon, are paid only. It sits in our Gemini vibe coding collection and you can see how it compares on our best vibe coding tools page. What Build mode does • Ways to start: describe an app in a prompt, press "I'm Feeling Lucky" for an idea, remix an app from the gallery, or import a GitHub project. • Web apps: generated apps use a React frontend and a Node.js server-side runtime. • Android apps: generated apps are Kotlin and Jetpack Compose projects that you can preview in a browser-based emulator. • Antigravity Agent: Google's docs describe it as the main AI functionality powering Build mode. • Full-stack services: Firebase is provisioned automatically, including Firestore and Google sign-in, and real-time multiplayer is listed as a capability. • Iterate: refine through chat or annotation mode, where you highlight part of the UI and describe a change. • GitHub sync: changes sync in both directions with a repository, with AI-generated commit messages and a conflict-resolution dialog. Deploy and export • Cloud Run: deploy as a scalable service. Google says Cloud Run pricing may apply. • GitHub: keep the source code in a repository. • ZIP download: export the code and run it yourself. You then set the GEMINI_API_KEY environment variable yourself. • API keys: on the web, the Gemini API key is set up as a server-side secret and is "never included in client-side code". Android apps use Firebase AI Logic and App Check instead of a raw key. • Paid models: the docs say paid models may incur costs when apps are shared. Pricing and free usage (as of October 2026) Google's Gemini API pricing page describes a free tier with "limited access to certain models", free input and output tokens and Google AI Studio access. The paid tier lists a Batch API with a 50% cost reduction. An enterprise route powered by Gemini Enterprise Agent Platform lists provisioned throughput and volume-based discounts. • Free of charge on the standard tier: Gemini 3.8 Flash and 3.6 Flash, Gemini 3.5 Flash-Lite and 3.1 Flash-Lite, several Live and transcription models. • Not on the free tier: Gemini 4 Argon, Gemini 3.1 Pro Preview, Gemini Omni Flash and the Nano Banana image models. Some can be tried inside Google AI Studio, and Google says Gemini 4 Argon is not available on the Free tier. • Data use: free-tier content is used to improve Google's products and paid-tier content is not. An invite-only Contributor Tier is an exception: its data is used to improve products. • Upcoming changes: the pricing page notes several rates changing from 1 December 2026 and 1 January 2027, so check it before budgeting. Related Google tools • Firebase Studio: Google's docs say it is sunsetting on 22 March 2027, new workspaces and sign-ups have been disabled since 22 June 2026, and existing workspaces can migrate to Google AI Studio or Google Antigravity. • Gemini CLI: its repository says free individual access and Google AI Pro and Ultra moved to Antigravity CLI on 18 June 2026. Who it suits • Developers who want a fast way to try Gemini models, prompts and API features in one place. • Builders who want a small working web app or Android app from a prompt, then a GitHub repository to continue in. • Teams already on Google Cloud who want to deploy to Cloud Run. • Firebase Studio users looking for a migration target. Limitations to know • Real-time collaborative editing is not supported. Teammates work through a shared GitHub repository. • Built-in storage support is described as planned. • Microphone, camera and similar device permissions must be declared in metadata.json. • The docs list content policy categories and an appeal process for flagged apps. • Browser privacy extensions or build issues can cause 403 errors when sharing. • Paid models and Cloud Run hosting carry separate costs. FAQs What is Google AI Studio? A Google workspace for building with the Gemini API. Build mode generates web and Android apps from prompts. Is Google AI Studio free? Google's Gemini API pricing page lists Google AI Studio access in the free tier, with free tokens on selected models. Newer models such as Gemini 4 Argon are not on the free tier. What does Build mode create? React and Node.js web apps, or Kotlin and Jetpack Compose Android apps, with a live preview. Where can I deploy an app? Cloud Run, a GitHub repository or a ZIP download. Is my API key exposed in the browser? For web apps Google says the Gemini API key is set up as a server-side secret and never included in client-side code. What we could not verify • Exact free-tier rate limits, which the pricing page sends to a separate page. • The main aistudio.google.com page shows only a title and tagline when read, so feature detail comes from Google's developer documentation. How this listing was put together This listing is based on Google AI Studio's own page, Google's Build mode documentation, the Gemini API pricing page, the Firebase Studio sunset notice and the Gemini CLI repository. Facts verified on 31 October 2026. Sources • Google AI Studio • Build mode documentation • Gemini API pricing • Firebase Studio sunset notice • Gemini CLI repository