Google AI Studio
Google AI Studio
Try Gemini and get an API key for free; better suited to people doing integrations than to everyday use.
Google's web environment for trying Gemini models, refining prompts, getting API keys and building AI apps from plain-language descriptions.
Overview
Google AI Studio is a browser-based environment for experimenting with Gemini models, adjusting prompts and system instructions, and getting sample code to integrate the Gemini API into your own product. According to the official pricing page, using Google AI Studio is free in supported regions.
Build mode lets you create apps from a description alone, often called vibe coding. By default AI Studio builds a web app with a React front end and a Node.js server, and it can also create Android apps using Kotlin and Jetpack Compose. Apps can be deployed to Cloud Run, synced with GitHub or downloaded as a ZIP.
For SMEs, this is a quick way to test AI app ideas, build prototypes for customers or get a Gemini API key for automation flows.
Highlights
Visual prompt testing
Adjust system instructions, model parameters and safety settings, then run immediately.
Build mode creates apps from descriptions
Describe an idea and AI Studio builds a web or Android app with a live preview.
Get integration code
The Get code button exports the prompt you are testing as Gemini API code in the language you choose.
Extended tools
Turn on structured output, function calling, code execution and grounding in Run settings.
Flexible deployment
Put the app on Cloud Run, sync with GitHub or download a ZIP to self-host.
What it's for
Prototyping AI apps
Quickly build a demo advisory chatbot or internal tool to present to customers or your boss.
Designing chatbot prompts
Try different system instructions so the chatbot answers in the brand's tone.
Getting a Gemini API key
Create a key to call Gemini from automation tools such as n8n or from apps you write yourself.
Learning and trying new models
Compare responses across Gemini models before choosing one for your product.
How to use it
- 1
Open AI Studio
Go to aistudio.google.com and sign in with a Google account.
- 2
Try a chat prompt
In the Playground, open Run settings to add a system instruction, enter a question and click Run to see the response.
- 3
Refine the instructions
Limit length, set a tone or role, then run again until it meets your requirements.
- 4
Get code or save the prompt
Use Get code to receive Gemini API code, or save and share the prompt with colleagues.
- 5
Build an app in Build mode
Describe the app you want, view the preview, keep refining via chat or edit directly in the Code tab.
- 6
Deploy
Put the app on Cloud Run, sync with GitHub or download a ZIP; if self-hosting, you must set the GEMINI_API_KEY environment variable yourself.
Video guides
Google AI Studio for Beginners
Google Cloud Tech
Building an Android app in AI Studio
Google for Developers
Google AI Studio: AI Miễn Phí Tốt Nhất Thay Đổi Mọi Cuộc Chơi (Cách Sử Dụng Google AI Studio)
Học Viện Marketing Online
Real-world example
A small English center in Can Tho wants a tool that generates topic-based vocabulary exercises for teachers. The center manager opens Google AI Studio and tries writing system instructions for Gemini until the exercises come out in the right format, then switches to Build mode and describes a simple web app with fields for topic and level. The prototype is shared with teachers to try, and once stable it is deployed to Cloud Run for the whole center to use.
Tips & notes
Write clear system instructions
Define the role, tone and length limits for more consistent responses.
Be careful with sensitive data
On the free tier, submitted content may be used to improve Google's products, so avoid entering real customer data.
Watch costs when sharing apps
API calls from a shared app count against your quota, and Cloud Run may also incur charges depending on usage.
Keep your API key secret
Do not put the key in browser-side code or post it on public GitHub.
FAQ
Is Google AI Studio free?
Using Google AI Studio is free in supported regions. The Gemini API has a limited free tier and paid tiers - see the official pricing page.
Can I use Build mode without knowing how to code?
You can start with descriptions alone, but before real use someone should still check the code and security.
How is AI Studio different from the Gemini app?
AI Studio is for trying models, designing prompts, getting API keys and building apps, rather than just everyday chat.
Where can the generated app be deployed?
It can be deployed to Cloud Run, synced with GitHub or downloaded as a ZIP to host elsewhere.
Useful links
Want to apply this in your business?
2dm looks at your workflows, picks the right tools and sets them up with your team. The first call is free.
Same topic
Claude Code
4.5
Anthropic's agentic coding assistant: reads your whole codebase, edits multiple files, runs commands and works with Git using plain-language instructions.
Cursor
4.5
A code editor with a built-in AI agent: understands your codebase, plans, writes features, fixes bugs and reviews changes right in the editor.
Google Antigravity
4.5
Google's agentic development platform: an app for running many agents in parallel, an IDE, a CLI (agy) and an SDK, powered by Gemini and Claude.