Google AI Studio — AI Coding Assistance

Google

Google AI Studio is Google's browser-based workspace for trying Gemini models, getting a Gemini API key and building apps. The Gemini API docs describe it as a place to test prompts, manage API keys, monitor usage and build prototypes, and the pricing page says AI Studio usage is free of charge in all available regions, while the billing FAQ adds that it stays free unless you link a paid API key. It is a developer product rather than the consumer Gemini app: the terms say it is for developers, for professional or business purposes and not for consumer use, and you must be 18 or older. The pricing page marks free-tier content as used to improve Google's products and paid-tier content as not, but the terms apply the paid-services data terms to AI Studio and unpaid quota for users in the European Economic Area, Switzerland and the United Kingdom. Per-model rate limits are not printed in the docs. Everything here is as read from Google's pages on 25 October 2026.

Why it fits

AI Coding AssistanceAI Chat AssistantsBuild an App with AI

Website

Go to aistudio.google.com
Good to know (4)

Check AI answers against their sources. Review account permissions and data settings before sharing private files or connections.

Check the official page for current quotas and output conditions.

Before publishing with user data or payments, review sign-in and database access rules, check whether the published app is public by default and keep secrets out of prompts and client code.

App Store and Google Play review, a developer account and testing on a real device are your responsibility.

Good to know
Good to knowWebsite
Free output—
No account needed—
No watermark—
Works offline—
Batch processing—
Content language—

— Unverified conditions. Official documentation reviewed. The app itself has not been hands-on tested. Matches only documented content-language support, not interface language.

Sources and review date (5)
  • ai.google.dev/gemini-api/docs/ai-studio-quickstart
  • ai.google.dev/gemini-api/docs/aistudio-build-mode
  • ai.google.dev/gemma/docs/core/deploy_to_cloud_run_from_ai_studio
  • ai.google.dev/gemini-api/docs/aistudio-build-mode
  • ai.google.dev/gemini-api/docs/ai-studio-quickstart
Founder feedback and correction

Founder feedback and correction

If you represent this product, send a source-backed correction or update to the editors. Publication follows editorial review.

Editorial selection guide

Last checked:

Who is it for?

  • Developers and technical users who want to try Gemini models and refine a prompt before writing code. The AI Studio quickstart describes a chat prompt with system instructions, a run settings panel (model parameters, safety settings, structured output, function calling, code execution and grounding) and a Get code button that turns the prompt into code for the Gemini API.
  • Anyone who needs a Gemini API key. The API key page says to click Create API key on the AI Studio API Keys page and that every key belongs to a Google Cloud project, and the Gemini API quickstart says AI Studio automatically creates a project and key for new users.
  • People who want to describe an app in plain language and get a working prototype. The Build mode page covers web apps (a React frontend with a Node.js server runtime) and Android apps (Kotlin and Jetpack Compose), with preview, GitHub import and two-way sync, ZIP export and deployment to Cloud Run.
  • Teams that want to prototype on the free tier and move to paid later. The pricing page lists a free tier with limited access to certain models and a paid tier with higher rate limits, context caching and a Batch API at a 50% cost reduction, and the billing page describes upgrading by linking Cloud Billing.

Getting started

  • Check access first. The troubleshooting page says you must be 18 or older to use Google AI Studio and must accept the Google Terms of Service and Generative AI Additional Terms. The regions page lists the countries and territories where the Gemini API and AI Studio are available and says your region may not be supported if you land on that page while trying to open AI Studio; the troubleshooting page names a 403 Access Restricted error as often a region problem.
  • Try a prompt in the Playground. The AI Studio quickstart says a new chat prompt opens, you add system instructions in Run settings, type a message and click Run. Every message in a chat is part of the prompt, so long conversations grow and may hit the model's token limit, and the troubleshooting page points to the Text Preview button to check a prompt's token count.
  • Reuse it as code or get a key. Click Get code and choose a language to carry the prompt into the Gemini API. To create a key, open the API Keys page and click Create API key. The key page says to treat it like a password, never commit it to source control and never hardcode it in web or mobile apps; it names GEMINI_API_KEY and GOOGLE_API_KEY as environment variables, and GOOGLE_API_KEY takes precedence if both are set.
  • To build an app, open Build mode and start from a prompt, the I'm Feeling Lucky button, an App Gallery project to remix or a GitHub import. Change it through the chat panel, the Code tab or annotation mode. The page says apps are private by default and the Gemini key is kept as a server-side secret, never in client-side code. Deploy to Cloud Run (the page says Cloud Run pricing may apply), sync with GitHub or export a ZIP, in which case you set GEMINI_API_KEY in your own hosting.
  • Check usage before you share anything. The Build mode page says API calls from a shared app count toward your usage limits and may cost money if it uses paid models, and that AI Studio warns during setup and before sharing. The billing page says usage is viewed in AI Studio under Dashboard then Usage, and the rate limits page says your active limits are shown in AI Studio.

Decision points

  • Free or paid: the pricing page lists the free tier as limited access to certain models, free input and output tokens and AI Studio access, and says AI Studio usage is free of charge in all available regions. The billing FAQ adds that it stays free unless you link a paid API key; once a paid key from a paid project is linked, you are charged for AI Studio usage on that key, and you can switch between Free Tier and Paid Tier projects by using each project's key. The paid tier lists higher rate limits for production deployments, context caching and a Batch API at a 50% cost reduction. Some models, such as the image generation models in its tables, show Not available on the free tier.
  • Data use differs by tier. The pricing page marks free-tier content as used to improve Google's products and paid-tier content as not. The terms add that for unpaid services human reviewers may read inputs and outputs after they are disconnected from your account, key and project, and that you should not submit sensitive, confidential or personal information. For paid services Google says it does not use prompts or responses to improve products and logs them for a limited period for abuse and legal reasons. The terms count AI Studio access as a paid service when the account has access to a Cloud project with active billing or is a Workspace enterprise account.
  • Moving to paid: the billing page says to click Set up billing on the API Keys or Projects page, link a Cloud Billing account and prepay at least $5 in credits; unused credits expire after 12 months, refunds are not allowed for Prepay accounts except when switching account types, and eligible accounts may be offered Postpay. Tier 1 starts when billing is linked, Tier 2 needs $100 of cumulative paid spend and 3 days since the first payment, and Tier 3 needs $1,000 and 30 days. Monthly billing-account caps are listed as $250, $2,000 and $20,000 to $100,000 or more (the page words the Tier 3 range slightly differently in two tables).
  • Understand the limits you will actually hit. The rate limits page says limits are measured in requests per minute, input tokens per minute and requests per day, apply per project and not per API key, and that daily request quotas reset at midnight Pacific time. It prints no per-model numbers, says specified limits are not guaranteed, and says preview and experimental models have tighter limits. Spend-based limits are listed as $10, $50 and $200 over a rolling ten minutes for Tiers 1 to 3.
  • Consumer or developer use: the terms say use of Google AI Studio and the Gemini API is for developers building with Google AI models, for professional or business purposes and not for consumer use, and that you must be 18 or older. The docs pages read do not compare AI Studio with the Gemini app, so use the Gemini app's own help pages for consumer questions, which this guide does not cover. For users in the European Economic Area, Switzerland and the United Kingdom, the terms apply the paid-services data terms to AI Studio and unpaid quota, and say you may use only Paid Services when making API clients available to users there.

Limits

  • The AI Studio app itself could not be read as a public page, so interface details here come from the documentation, not from the app. The rate limits page prints no per-model numbers and points to the rate limit view inside AI Studio, and the pricing page lists future price changes with dates (December 2026 and January 2027 for some models). Check the signed-in interface and the live pricing page before planning around a number.
  • Data use in Europe works differently from the pricing table. The terms say that if you are in the European Economic Area, Switzerland or the United Kingdom, the paid-services data terms apply to all Services, including Google AI Studio and unpaid quota in the Gemini API, even though they are offered free of charge. The pricing table's free-tier row (content used to improve products) therefore does not describe users there, and the pricing page itself carries no note about this. Read the terms for your own location before relying on either rule.
  • Build mode has no preview or experimental label on the page and no Build-specific price or quota figures; Cloud Run pricing may apply, and the page does not say whether billing is required to deploy. It says real-time collaborative editing is not supported and collaboration happens through a shared GitHub repository, that apps flagged for policy violations may be removed, and that Android apps can be published to the Play Store for internal testing.
  • The documentation is moving. The docs home says the Interactions API became the default interface as of June 2026 and that generateContent is now considered legacy. The logs page says Interactions requests are stored by default while generateContent requests are not, that log storage is available only to paid-tier projects, and that the default retention is 55 days. The docs pages read show last-updated dates between July and October 2026 and the terms are dated April 2026, so older tutorials may not match.
  • This guide summarizes Google's pages as read on 25 October 2026. It is not a hands-on test and gives no rating. Prices are in US dollars and can change, the region list is long and changes, and a Workspace for Education account depends on whether the administrator has designated it as 18 or older.

Frequently asked questions

Is Google AI Studio free to use?

The pricing page says Google AI Studio usage is free of charge in all available regions, and lists the free tier as limited access to certain models, free input and output tokens and AI Studio access. The billing FAQ adds that this holds unless you link a paid API key: once a paid key from a paid project is linked, AI Studio usage on that key is charged. Some models, including the image generation models in the pricing tables, show Not available on the free tier. The paid tier lists higher rate limits, context caching and a Batch API at a 50% cost reduction. Check the live pricing page for the model you want.

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

Open the API Keys page in AI Studio and click Create API key. The key page says new users get a default project and key after accepting the Terms of Service, that people with existing Google Cloud projects may need to import them from the Dashboard first, and that creating a key needs IAM permissions such as apikeys.keys.create. Every key is tied to a Google Cloud project, and the page says to keep it out of source control and client-side code and to use GEMINI_API_KEY or GOOGLE_API_KEY as the environment variable.

Is Google AI Studio the same as the Gemini app, and who can use it?

The docs pages read describe AI Studio as a place to test prompts, manage API keys, monitor usage and build prototypes for the Gemini API, and they do not compare it with the Gemini app. The terms say it is for developers, for professional or business purposes and not for consumer use. The troubleshooting page says you must be 18 or older and the regions page lists the countries and territories where it is available. Use the Gemini app's own help pages for the consumer app, which this guide does not cover.

Does Google use my Google AI Studio prompts to train its models?

It depends on the tier and on where you are. The pricing page marks free-tier content as used to improve Google's products and paid-tier content as not. The terms say that for unpaid services Google uses submitted content and responses to improve its products, human reviewers may read inputs and outputs after they are disconnected from your account, key and project, and you should not submit sensitive, confidential or personal information. For paid services, meaning API use through a Cloud project with an active billing account, Google says it does not use prompts or responses to improve its products and logs them for a limited period to detect policy violations. If you are in the European Economic Area, Switzerland or the United Kingdom, the terms apply the paid-services data terms to AI Studio and unpaid quota too.

Can Google AI Studio build apps, and what does that cost?

Yes, the Build mode page describes it. You describe an app in plain language and it builds a web app with a React frontend and Node.js server runtime or an Android app in Kotlin and Jetpack Compose, which you can edit in chat or in the Code tab, sync with GitHub, export as a ZIP or deploy to Cloud Run. The page gives no Build-specific price: Cloud Run pricing may apply, and API calls from a shared app count toward your usage limits and may cost money if it uses paid models. Apps are private by default.

Is there a review of Google AI Studio, and what should I compare if I want an alternative?

OOXL has not tested Google AI Studio and gives no rating. Google's own pages let you check what matters: AI Studio usage is free of charge in available regions with limited model access unless you link a paid API key; outside the European Economic Area, Switzerland and the United Kingdom, free-tier content is used to improve Google's products; you must be 18 or older; the terms are written for professional or business use; and per-model rate limits are shown inside AI Studio, not in the docs. When comparing another tool for prompting models, getting an API key or building apps, check the same points on that tool's own pages: cost, data use on the free tier, age and region rules, published limits and whether it exports your work.

Official sources (12)
ai.google.dev/gemini-api/docsai.google.dev/gemini-api/docs/ai-studio-quickstartai.google.dev/gemini-api/docs/aistudio-build-modeai.google.dev/gemini-api/docs/api-keyai.google.dev/gemini-api/docs/pricingai.google.dev/gemini-api/docs/rate-limitsai.google.dev/gemini-api/docs/billingai.google.dev/gemini-api/termsai.google.dev/gemini-api/docs/available-regionsai.google.dev/gemini-api/docs/troubleshoot-ai-studioai.google.dev/gemini-api/docs/logs-datasetsai.google.dev/gemini-api/docs/quickstart