Gemini 2.5 Flash Image (Nano Banana)

Overview

Gemini 2.5 Flash Image, also known as Nano Banana, is our fastest image generation engine, designed for speed and efficiency. Optimized for high-volume, low-latency tasks, Nano Banana generates images at up to 1024px resolution with an excellent quality-to-speed ratio. This makes it the ideal choice for applications where response time and cost matter more than maximum resolution.

Best for: high-volume generation, conversational image editing, rapid prototyping, and low-latency creative workflows where budget efficiency is a priority.

Model Variants

The standard Nano Banana variant costs only 2 credits per image, making it the most affordable option in our entire model lineup. The Nano Banana HD variant enhances detail and sharpness while retaining the same resolution, ideal when you need sharper results without scaling up to a higher-tier model.

Capabilities

Supported Aspect Ratios

1:1 · 16:9 · 9:16 · 4:3 · 3:4 · 2:3 · 3:2 · 4:5 · 5:4

Pricing

All pricing is based on per-image cost via the NanoBanana API, significantly cheaper than official channels.

At nearly half the cost of official pricing, Nano Banana provides the best cost-per-image ratio in the lineup. For budget-conscious teams processing thousands of images, the savings add up significantly over time.

Performance Characteristics

Nano Banana Speed

Nano Banana is the fastest engine in our lineup, typically returning results in under 2 seconds for standard resolution requests. The lower computational overhead means:

• Lower latency: Ideal for real-time or interactive applications where users expect near-instant feedback
• Higher throughput: Process more concurrent requests without hitting rate limits as quickly
• Faster iteration cycles: Designers and developers can experiment with prompts and see results quickly

Quality Trade-offs

While this engine does not support Thinking mode or Search Grounding, it produces excellent results for many common use cases. The quality is comparable to competing services at the 1K resolution tier, making it suitable for:

• Social media content (Instagram posts, Twitter cards, Facebook ads)
• Blog and article illustrations
• Thumbnails and preview images
• Chat and messaging sticker creation
• Placeholder imagery for UI prototypes

For tasks requiring precise text rendering, complex compositions, or factual accuracy, consider upgrading to the Pro tier instead.

Best Practices

Maximizing Quality on a Budget

• Use the HD variant for important assets: When output quality matters but you still need to keep costs low, the HD variant at 5 credits provides noticeably sharper results than the standard 2-credit option.
• Keep prompts focused: Since this engine does not have Thinking mode, simpler and more focused prompts tend to produce better results than long, multi-subject descriptions.
• Leverage batch generation: Submit up to 9 images per API call and pick the best result. Even at 9× the base cost, the total price remains very competitive.
• Iterate quickly: Use the low cost to your advantage — generate multiple variations and refine your prompt based on the results.

When to Upgrade

Consider switching to a higher-tier model when your project requires:

• Output resolution above 1024px (use the Flash 3.1 tier for up to 4K)
• Complex scenes with precise spatial relationships (use the Pro tier with Thinking)
• Accurate depictions of real-world subjects (use the Pro tier with Search Grounding)
• Extended aspect ratios like 21:9 or 1:8 (exclusive to the Flash 3.1 tier)

Use Cases

• Rapid prototyping — Quickly iterate on visual concepts during brainstorming and design sprints
• High-volume batch generation — Generate large sets of images cost-effectively for product catalogs or content libraries
• Conversational editing — Interactive image refinement workflows where multiple rounds of editing are expected
• Real-time applications — Low-latency generation for user-facing products such as chatbots and creative tools
• A/B testing creatives — Generate multiple ad or banner variations cheaply to test performance
• Content automation — Automated pipelines that produce visuals for articles, social posts, or email campaigns at scale
• Sticker and emoji creation — Simple, stylized imagery for messaging platforms and community tools

Quick Start

Replace YOUR_API_KEY with your actual API key. Don't have one yet? Create an API key here.

curl -X POST "https://api.nanobananaapi.dev/v1/images/generate" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A cute cat sitting on a windowsill watching the rain",
    "num": 1,
    "model": "gemini-2.5-flash-image",
    "image_size": "1:1"
  }'

const res = await fetch('https://api.nanobananaapi.dev/v1/images/generate', {
  method: 'POST',
  headers: {
    Authorization: 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    prompt: 'A cute cat sitting on a windowsill watching the rain',
    num: 1,
    model: 'gemini-2.5-flash-image',
    image_size: '1:1',
  }),
});

const result = await res.json();
console.log(result.data.url);

import requests

res = requests.post(
  'https://api.nanobananaapi.dev/v1/images/generate',
  headers={
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  json={
    'prompt': 'A cute cat sitting on a windowsill watching the rain',
    'num': 1,
    'model': 'gemini-2.5-flash-image',
    'image_size': '1:1',
  },
  timeout=60,
)

result = res.json()
print(result['data']['url'])

API Parameters Reference

Frequently Asked Questions

Why does Nano Banana not support Thinking or Search Grounding? These features require additional computational resources that would increase latency and cost. By omitting them, this engine can deliver the fastest response times and lowest per-image pricing in the lineup. For most standard image generation tasks, these features are not necessary.

Is the HD variant worth the extra cost? If your images will be viewed at full size or on high-DPI displays, the HD variant provides noticeably better detail and sharpness. For thumbnails, previews, or small-format usage, the standard variant is usually sufficient and costs less than half.

Can I use this model for image editing? Yes. Upload up to 3 reference images along with a text prompt describing the desired changes. The model supports conversational editing, allowing multiple rounds of refinement.

What happens if I need higher resolution later? You can always re-generate your images using a higher-resolution model. Consider saving your prompts and parameters so you can reproduce results at a higher tier when needed.

Related

• Text to Image API — Full API reference for image generation
• Image to Image API — Image editing and transformation
• Gemini 3.1 Flash Image Preview — Higher resolution with extended aspect ratios
• Gemini 3 Pro Image Preview — Professional-grade with Thinking and Search Grounding