Google AI Studio (2026): What It Is and How to Use It

By Antigravity Skills editorial team · Last updated

Short answer: Google AI Studio is Google's browser workspace for trying Gemini models and building apps with them. You open aistudio.google.com, sign in with a Google account, test prompts, or describe an app in Build mode and let it write the code. It also issues the Gemini API key your own code and coding agents use. Google lists a free tier, so you can start without paying.

Skill page: Google publishes official Gemini agent skills for coding agents, covered in Gemini API agent skills: install and use, with the install commands and what each skill gives your agent.

Unofficial guide. Antigravity Skills is not affiliated with Google. The official source is aistudio.google.com and its docs. Facts below were checked on November 23, 2026 against Google's AI Studio documentation and the google-gemini/gemini-skills repository. Limits, prices and model names change often, so confirm them at the source.

At a glance

What it isBrowser workspace for prompting Gemini models, building apps and getting API keys
Addressaistudio.google.com
Sign-inA Google account
Build modeDescribe an app in plain language; AI Studio writes the code and shows a live preview
What Build mode makesWeb apps (React front end, Node.js server) or native Android apps (Kotlin and Jetpack Compose)
API keyCreated in AI Studio and used with the Gemini API
PriceGoogle's pricing page lists a free tier and paid usage; read the current page before you rely on a number
Agent skillsOfficial Gemini skills exist and come pre-installed in AI Studio Build

What is Google AI Studio?

It is the place where Google lets developers and curious users work with Gemini directly. Google's documentation describes it as the way to quickly build, or "vibe code", and deploy apps that test the latest Gemini features, such as image generation and the Live API. There are three things people use it for:

  • Prompt testing. Try a prompt against a Gemini model, change the settings and compare the answers before you write code.
  • Build mode. Turn a text description into a working app with a live preview.
  • API keys. Create the key that your own program, or a coding agent, uses to call the Gemini API.

How to use Google AI Studio

  1. Open aistudio.google.com and sign in with a Google account.
  2. Pick a model and type a prompt to test it. Change the settings and run it again until the answer is what you want.
  3. To build an app, open Build mode and describe it. You can also press the "I'm Feeling Lucky" button for a starter idea, copy an app from the App Gallery, or import a project from GitHub.
  4. Read the generated files in the Code tab, and ask for changes in the prompt box. The preview updates next to it.
  5. When you need the model inside your own program, create an API key in AI Studio and keep it secret.

What does Build mode create?

Per Google's Build mode documentation, you choose a web app or a native Android app in the platform picker.

  • Web apps get a full-stack environment: a React front end by default and a Node.js server that can make secure API calls, connect to databases and use npm packages.
  • Android apps get a Kotlin and Jetpack Compose project that you can preview in a browser emulator, install on a device and publish to the Play Store for testing.
  • Google says the Antigravity Agent manages several files across the stack so a change in one place carries through.
  • For web apps, Google says AI Studio sets your Gemini API key as a server-side secret, so it is not exposed in the browser.

Is Google AI Studio free?

Google's pricing page lists a free tier with free input and output tokens and limits, and paid usage beyond it. Not every model is on the free tier, and the page warns that some tools such as search grounding are billed separately above a monthly allowance. Prices change, so check Google's pricing page on the day you build.

Google AI Studio and Antigravity

The names overlap because Google uses both. AI Studio is the browser workspace; Antigravity is Google's agent coding tool. They connect in two ways in Google's docs:

  • Build mode's code agent is called the Antigravity Agent.
  • Google's Gemini skills are bundled with Antigravity. The repository README says you find them under Customizations, or install them with agy plugin install https://github.com/google-gemini/gemini-skills.

On this site, Antigravity Skills is the library of agent skills that work in Antigravity, Claude Code and other agents. If you are new to the format, read What are Antigravity Skills? and how to install skills.

Use agent skills with Google AI Studio

A coding agent that writes Gemini code from memory often suggests old models or the old google-generativeai SDK. Google's coding-agent page says connecting official resources lets the agent use current models, the google-genai SDK and the Interactions API. Four skills are in the official repository:

SkillWhat it covers
gemini-api-devBuilding apps with the Gemini API: text, chat, streaming, function calling, structured output, images, speech, agents
gemini-api-cliCalling the Gemini API from the shell with the gemini-api command-line tool
gemini-live-api-devReal-time streaming apps with the Live API
gemini-omni-flash-apiVideo generation and editing with the Omni Flash model

Install and usage details are on Gemini API agent skills: install and use. Skills come pre-installed in AI Studio Build, per Google.

Limits to know

  • Privacy. Read Google's terms for the free tier before you paste private data into a prompt. Free and paid tiers can be handled differently.
  • Keys. Never put an API key in a public repository or front-end code.
  • Fast changes. Model names, limits and Build mode features change often. This page is a snapshot.
  • We did not build an app ourselves. This guide is from Google's documentation and the skills repository, not from a hands-on test.

FAQ

What is Google AI Studio?

It is Google's browser workspace at aistudio.google.com for testing Gemini models, building apps in Build mode and creating Gemini API keys.

Is Google AI Studio free?

Google lists a free tier, with limits, plus paid usage. Check the current pricing page, because limits and eligible models change.

How do I get a Gemini API key in Google AI Studio?

Sign in at aistudio.google.com and create a key from the API keys area. Keep it secret and read Google's current steps in the Gemini API documentation.

What is the difference between Google AI Studio and Gemini?

Gemini is the family of models, and the Gemini app is the consumer chat. AI Studio is the developer workspace where you use those models, build apps and get API keys.

What is Build mode?

It is the part of AI Studio where you describe an app and get generated code with a live preview, for the web or for Android.

Does Google AI Studio have agent skills?

Yes. Google publishes official Gemini skills in google-gemini/gemini-skills, and they come pre-installed in AI Studio Build.

Sources