How to Use Google AI Studio
Prototype AI features with Gemini — free, fast, and multimodal.
🌐Browser — aistudio.google.com — sign in with Google, free to useGoogle AI Studio is a free browser tool for experimenting with Google's Gemini AI models. You can build AI-powered features — text summarizers, image analyzers, chatbots — directly in the browser and export working code when you're ready to build a full app around it.
✓ Perfect for you if...
- ▸Prototyping AI features before building a full app around them
- ▸Apps that need to analyze images, audio, or long documents
- ▸Anyone already in the Google ecosystem (Workspace, Firebase, Cloud)
Get set up in 5 steps
Follow these in order — each step takes about 1–2 minutes.
Go to aistudio.google.com
Open aistudio.google.com in your browser.
Sign in with your Google account
Click "Sign in." Use the same Google account you use for Gmail or YouTube — no separate signup needed.
Click "Create new prompt"
On the left sidebar, click "Create new prompt." This opens the prompt playground where you can experiment with Gemini.
Choose a model
At the top of the page, select a Gemini model. "Gemini 2.0 Flash" is the fastest and most capable free model — start here.
Type your prompt and click "Run"
Write what you want Gemini to do and press the Run button (or Ctrl+Enter). Try asking it to summarize text, answer questions, or analyze an image you upload.
Your first prompt — paste this right now
This is a "System Prompt" — paste it in the System Instructions box in AI Studio. It turns Gemini into a specialized email assistant. Click "Get code" to export this as a working Python or JavaScript snippet.
You are an email assistant. I will paste a long email thread below and you will: 1. Summarize the key points in 3 bullet points 2. Identify any action items I need to take (as a separate bullet list) 3. Suggest a short, professional reply I can send Keep the summary under 100 words. Be direct and practical. [Paste your email thread here]
3 tips that make a real difference
1. Use the "System Instructions" box for reusable behavior
The System Instructions box (above the chat) defines how Gemini behaves for every message. Use it to set a role ("You are a helpful recipe assistant"), rules, or output format. This is how you turn Gemini into a specialized tool.
2. Click "Get code" to export a working implementation
Once your prompt works the way you want, click the "</> Get code" button to export it as Python, JavaScript, or a curl command. This is the bridge between experimenting in AI Studio and building a real app.
3. Try uploading an image or document
Gemini is multimodal — click the attachment button to upload a photo, PDF, or audio file and ask questions about it. This is one of Gemini's biggest advantages over other models.
Ready to pick your first idea?
We've curated the best project ideas specifically matched to Google AI Studio — filtered by what actually works well with this tool.
Browse Google AI Studio project ideas →Also learn how to use