Skip to main content
Google AI Studio provides free access to Gemini and Gemma models for prototyping and development.

​
Overview

Google AI Studio offers free API access to Google’s latest Gemini and Gemma models with generous rate limits suitable for development and experimentation.
Data Usage Policy: Data is used for training when used outside of the UK, Switzerland, EEA, or EU.

​
Rate Limits

Each model has specific rate limits:
Model NameTokens/MinuteRequests/DayRequests/Minute
Gemini 3 Flash250,000205
Gemini 3.1 Flash-Lite250,00050015
Gemini 2.5 Flash250,000205
Gemini 2.5 Flash-Lite250,0002010
Gemma 3 27B Instruct15,00014,40030
Gemma 3 12B Instruct15,00014,40030
Gemma 3 4B Instruct15,00014,40030
Gemma 3 1B Instruct15,00014,40030

​
Available Models

​
Gemini Models

Gemini 3 Flash

Fast, efficient model with 250K tokens/minute

Gemini 3.1 Flash-Lite

Lightweight variant with higher request limits

Gemini 2.5 Flash

Previous generation Flash model

Gemini 2.5 Flash-Lite

Lightweight 2.5 generation model

​
Gemma Models

All Gemma models share the same rate limits: 15,000 tokens/minute, 14,400 requests/day, and 30 requests/minute.
  • Gemma 3 27B Instruct - Largest open model
  • Gemma 3 12B Instruct - Balanced performance
  • Gemma 3 4B Instruct - Efficient smaller model
  • Gemma 3 1B Instruct - Ultra-compact model

​
API Usage

import google.generativeai as genai

genai.configure(api_key="YOUR_API_KEY")

model = genai.GenerativeModel('gemini-2.5-flash')
response = model.generate_content('Hello, how are you?')

print(response.text)

​
Getting Started

1

Visit Google AI Studio

Go to aistudio.google.com
2

Sign in with Google

Use your Google account to sign in
3

Generate API Key

Create an API key from the dashboard
4

Start Building

Use the API key to make requests to Gemini and Gemma models

​
Privacy Considerations

If you’re using Google AI Studio from outside the UK, Switzerland, EEA, or EU, be aware that your data may be used for training purposes. For production applications with stricter privacy requirements, consider using Google Cloud Vertex AI instead.

​
Additional Resources

Google AI Studio

Access the web interface

Documentation

Official API documentation

Build docs developers (and LLMs) love

Get started for freeTalk to us