api-access-guide

Purpose

Complete guide to accessing and using Google’s Nano Banana and Nano Banana Pro image generation APIs, including setup, authentication, pricing, and capabilities.

What is Nano Banana?

“Nano Banana” is the community nickname for Google’s Gemini image generation models:

Model Versions

Name	Official Name	Model ID	Best For
Nano Banana	Gemini 2.5 Flash Image	`gemini-2.5-flash-image`	Fast, cost-effective generation
Nano Banana Pro	Gemini 3 Pro Image	`gemini-3-pro-image-preview`	High quality, advanced features

Relationship to NotebookLM

Important distinction: While NotebookLM’s Video Overview feature uses “Gemini Nano Banana” for image generation, that is NOT the same as these standalone APIs. NotebookLM’s Video Overview has no API access - it’s web UI only.

These standalone Gemini image APIs are available to all developers through Google AI Studio.

How to Get API Access

Method 1: Google AI Studio (Recommended)

Step-by-step setup:

Create Google Cloud Account
- Visit Google AI Studio
- Sign in with Google account
Get API Key
- Click “Get API key” button in left sidebar
- Generate new API key or use existing project
Enable Billing (Required for Nano Banana Pro)
- Nano Banana Pro has no free tier
- Navigate to Google Cloud Console
- Enable billing on your project
Install SDK (Optional)
- Python: pip install google-generativeai
- JavaScript: npm install @google/generative-ai
- Also available: Java, Go, REST
Start Making API Calls
- Use model identifiers: gemini-2.5-flash-image or gemini-3-pro-image-preview

Method 2: Vertex AI (Enterprise)

Access through Google Cloud’s Vertex AI platform:

Better for enterprise deployments
Integrates with existing GCP infrastructure
Same models, different access layer

Method 3: Third-Party Platforms

Unified API Access:

Kie.ai - Unified API key for both Nano Banana and Nano Banana Pro
CometAPI - All AI models in one API
Puter.js - Frontend access without server setup

Benefits of third-party platforms:

Single API key for multiple providers
Simplified billing
Additional features (playground, model comparison)
Often provide free tiers or trials

Pricing

Official Google Pricing (2025)

Tier	Access	Cost	Limits
Free (2.5 Flash)	Google AI Studio API	Free	500 requests/day per project
Paid (2.5 Flash)	Billing enabled	~$0.15 per 4K gen	Usage-based
Gemini App (Consumer)	Web UI	Free	3 low-res generations, then degraded quality
Nano Banana Pro	Billing required	~$0.15 per 4K gen	No free tier

Special Promotions

48-hour free tier: Launched Sept 6, 2025 - 500 requests/day for free tier API keys
Check Google AI Studio for current promotions

Resolution Pricing

Low-res: Often free in Gemini app (first 3 generations)
2K output: Requires paid plan or API billing
4K output: ~$0.15 per generation (highest quality)

API Capabilities

Core Features

1. Text-to-Image Generation

Prompt with natural language
Generate images from scratch
Support for detailed descriptions

2. Image Editing

Modify existing images
Style transfers
Object addition/removal

3. Multi-Image Composition

Combine multiple images
Create collages
Blend styles and elements

4. High-Fidelity Text Rendering

Accurate text generation in images
Legible and well-placed text
Perfect for:
- Logos
- Diagrams
- Posters
- Infographics

Input Methods

You can prompt with:

Text only: Traditional text-to-image
Image only: Image-to-image editing
Text + Image: Combined prompts for precise control
Multiple images: Multi-image composition

Output Options

Resolutions: Up to 4K
Formats: Standard image formats (PNG, JPEG)
Iterations: Create, edit, and iterate on visuals

Nano Banana vs Nano Banana Pro

Feature Comparison

Feature	Nano Banana (2.5 Flash)	Nano Banana Pro (3 Pro)
Speed	⚡ Faster	Slower (higher quality processing)
Cost	💰 Lower	Higher
Quality	Good	⭐ Excellent
Free Tier	✅ 500 req/day	❌ No free tier
Best For	Rapid prototyping, high-volume	Professional work, maximum quality
Resolution	Up to 4K	Up to 4K
Text Rendering	Good	Excellent

When to Use Which

Use Nano Banana (2.5 Flash) when:

Prototyping and experimentation
High-volume generation needs
Cost is a primary concern
Speed matters more than perfect quality
You want to stay in free tier

Use Nano Banana Pro (3 Pro) when:

Professional/commercial projects
Highest quality is essential
Complex compositions required
Text accuracy is critical
Budget allows for premium quality

SDK Integration

Python Example

import google.generativeai as genai

# Configure API key
genai.configure(api_key="YOUR_API_KEY")

# Initialize model
model = genai.GenerativeModel('gemini-2.5-flash-image')

# Generate image
response = model.generate_content([
    "A futuristic cityscape at sunset with flying cars"
])

# Access generated image
image = response.images[0]

JavaScript Example

import { GoogleGenerativeAI } from "@google/generative-ai";

// Initialize client
const genAI = new GoogleGenerativeAI("YOUR_API_KEY");

// Get model
const model = genAI.getGenerativeModel({ model: "gemini-2.5-flash-image" });

// Generate image
const result = await model.generateContent([
  "A futuristic cityscape at sunset with flying cars"
]);

// Access generated image
const image = result.response.images[0];

REST API Example

curl https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "contents": [{
      "parts": [{
        "text": "A futuristic cityscape at sunset with flying cars"
      }]
    }]
  }'

Developer Resources

Official Documentation

Image generation with Gemini - Official API docs
Google AI Studio - Playground and API key management
Gemini 3 Pro Image Developer Guide - Official announcement

Community Resources

Nano Banana Hackathon Kit - Official starter kit
Complete Developer Tutorial - DEV Community guide
Step-by-Step Guide for Developers - Third-party guide

Alternative Access Platforms

Free Access Options

1. Google AI Studio Free Tier

Model: Gemini 2.5 Flash Image only
Limit: 500 requests/day per project
Quality: Full quality (up to 4K)
Best for: Personal projects, testing

2. Gemini App (Consumer)

Access: Web UI at gemini.google.com
Limit: 3 low-res generations
After limit: Falls back to lower quality models
Best for: Casual experimentation

3. Puter.js User-Pays Model

Setup: No server required, frontend-only
Cost: Users cover their own usage
Best for: Web apps where users pay per use

4. Third-Party Free Trials

Many platforms offer initial free credits
Check Kie.ai, CometAPI for current promotions

Migration from NotebookLM

If you were looking for NotebookLM Video Overview API:

NotebookLM’s Video Overview feature (which uses Gemini Nano Banana internally) does not have API access. It’s web UI only.

Alternatives:

Use these standalone APIs: Build your own video generation pipeline
- Generate images with Nano Banana API
- Combine with Google TTS API for narration
- Use ffmpeg for video composition
NotebookLM Enterprise API: Available for audio overviews only (not video)
- Podcast API: podcasts method
- Audio Overview: notebooks.audioOverviews.create
Third-party alternatives:
- AutoContent API - NotebookLM alternative
- Podcastfy - Open source podcast generation

Common Use Cases

Professional Applications

Marketing: Social media graphics, ad creatives
Design: Logo concepts, poster designs
Education: Diagrams, infographics, visual aids
Content Creation: Blog illustrations, thumbnails
Product Design: Mockups, concept art

Developer Applications

Automated content generation: Bulk image creation
User-generated content: Let users create custom visuals
Data visualization: Transform data into visual stories
Game assets: Generate textures, sprites, concepts
Prototyping: Rapid UI mockup generation

Limitations and Considerations

Current Limitations

No free tier for Pro version - Billing required for Gemini 3 Pro Image
Rate limits - 500 req/day on free tier
Resolution caps - Maximum 4K output
Content policies - Subject to Google’s AI usage policies

Best Practices

Start with free tier - Test with Gemini 2.5 Flash Image first
Optimize prompts - Better prompts = better results, fewer retries
Use appropriate model - Don’t pay for Pro if Flash suffices
Monitor usage - Track API calls to avoid unexpected costs
Cache results - Store generated images to avoid regeneration

Support and Community

Getting Help

Google AI Developers Forum - Official support forum
Gemini API Discord - Community chat
Stack Overflow - Tagged questions

Reporting Issues

GitHub Issues - SDK bugs and feature requests
Google Cloud Support - Enterprise support

Summary

Topic	Key Points
Access	Google AI Studio, Vertex AI, or third-party platforms
Models	Nano Banana (2.5 Flash) for speed, Pro (3 Pro) for quality
Free Tier	500 req/day for 2.5 Flash via Google AI Studio
Pricing	~$0.15 per 4K generation for paid usage
Capabilities	Text-to-image, editing, composition, text rendering
SDKs	Python, JavaScript, Java, Go, REST
Best For	Professional image generation, automated content creation
NotebookLM	Separate product - Video Overview has no API