Nano Banana
Model Family
Google DeepMind's complete image generation lineup — from studio-grade Pro and the new 2.1 to the family's fastest, Lite

Prompt
Create an infographic explaining how coffee is brewed, with labeled steps in a warm illustration style...
Four models, One API
Pick the Nano Banana that fits your speed, quality and budget — all on one unified image API, switched by a single parameter.
Nano Banana Pro
Google's flagship image model (Gemini 3 Pro Image) for studio-quality creative work and professional text rendering.
- Resolution
- 1K · 2K · 4K
- Aspect Ratios
- 10
- Reference Images
- 8
- Price
- from $0.06
Print design, infographics, storyboards and text-heavy visuals.
View Nano Banana ProNano Banana 2.1
The next generation of Nano Banana 2 (gemini-nano-banana-2.1): more realistic, better layout and text rendering, steadier characters across edits — at a lower price.
- Resolution
- 1K · 2K · 4K
- Aspect Ratios
- 10
- Reference Images
- 14
- Price
- From $0.02
E-commerce shots, poster layouts, multi-turn edits and character-consistent series.
Explore Nano Banana 2.1Nano Banana 2
Pro-level quality at Flash speed (Gemini 3.1 Flash Image) — the everyday workhorse with exclusive banner ratios.
- Resolution
- 512px · 1K · 2K · 4K
- Aspect Ratios
- 14
- Reference Images
- 14
- Price
- from $0.04
Marketing assets, greeting cards, banners and fast iteration.
View Nano Banana 2Nano Banana 2 Lite
The family's fastest, most cost-efficient model (Gemini 3.1 Flash Lite Image) — 1K generation in about 4 seconds, built for real-time products.
- Resolution
- 1K
- Aspect Ratios
- 10
- Reference Images
- 14
- Price
- $0.02
Stickers, recoloring, background swaps and high-concurrency apps.
View Nano Banana 2 LiteNano Banana Model Comparison
Same API, same async task flow — different speed, quality and price points.
Nano Banana Pro Studio Flagship | Nano Banana 2.1 Next-gen · Lower price | Nano Banana 2 Pro-Level, Flash Speed | Nano Banana 2 Lite Ultra-Fast & Low-Cost | |
|---|---|---|---|---|
| Positioning | Studio flagship | Next-gen · Lower price | Pro-level, Flash speed | Ultra-fast, low-cost |
| Upstream Model | Gemini 3 Pro Image | gemini-nano-banana-2.1 | Gemini 3.1 Flash Image | Gemini 3.1 Flash Lite Image |
| Resolution | 1K · 2K · 4K | 1K · 2K · 4K | 512px · 1K · 2K · 4K | 1K |
| Aspect Ratios | 10 standard | 10 standard (banner ratios coming) | 10 standard + 4 banner (1:4, 4:1, 1:8, 8:1) | 10 standard |
| Reference Images | 8 | 14 | 14 | 14 |
| Subject Consistency | Up to 5 people | 4 characters · 10 objects, across edit turns | 5 characters · 14 objects | Strong character consistency |
| Text Rendering | Best-in-class, multilingual | Improved layout and text rendering | Legible text + localization | Legible in-image text |
| Speed | Highest quality, slower | Fast (Flash tier) | Fast (Flash tier) | Fastest (~4s per image) |
| Price / Image | $0.06 · $0.09 · $0.12 (1K/2K/4K) | $0.02 (1K) · $0.04 (2K) · $0.05 (4K) | $0.04 (512px/1K) · $0.06 (2K) · $0.08 (4K) | $0.02 (1K) |
| Best For | Professional design & print | Everyday production, for less | Everyday production workhorse | Real-time, high-concurrency |
| Google Search Grounding |
All four models support text-to-image and image editing, are billed per generated image, and share the same async task API.
Real Use Cases, Real Outputs
Official examples from Google's Nano Banana launches — copy a prompt and try it yourself.
When quality and text accuracy are non-negotiable.

Knowledge Infographics
Create an infographic about this plant focusing on interesting information.

Film Storyboards
Create a storyboard for this scene.

Localized Campaigns
Translate all the English text on the three yellow and blue cans into Korean, while keeping everything else the same.
The daily driver for marketing and content teams.
Posters & Mockups
A candid photograph captures a vibrant row of freshly wheat-pasted posters adhered to a heavily textured, dark green urban wall in Downtown Los Angeles. The posters are for an "Indie Film Fest," with "DOWNTOWN LA" printed in smaller letters directly below. The bold design alternates across three distinct color palettes: one with an electric purple background and orange and white text, the next a blazing orange background with lime green and white text, and the third a lime green background with purple and white text, creating a pulsating visual rhythm down the line.
14-Reference Consistency
Create an image of these 14 characters and items having fun at the farm. The overall atmosphere is fun, silly and joyful. It is strictly important to keep identity consistent of all the 14 characters and items.
Character-Consistent Stories
Create a funny 6 part story with these 3 fluffy friends building a tree house. The story is thrilling throughout with emotional highs and lows and is ending in a happy moment. Keep the attire and identity consistent of all 3 characters, but their expressions and angles should vary throughout all 6 images.
Built for instant, high-volume interactions.
Lite vs 2: Speed Test
Watch Google's official side-by-side: Nano Banana 2 Lite and Nano Banana 2 generate the same prompt — Lite lands first, at production quality.

Quality × Speed × Price
Google's official benchmark of the Nano Banana family: Elo quality against latency and cost per 1K image, where Lite owns the fast-and-cheap corner.
Images and video are from Google's official Nano Banana announcements (blog.google, deepmind.google).
Integrate in Seconds
One unified API covers the whole Nano Banana family — switch between Pro, 2.1, 2 and Lite by changing a single model parameter.
import requests import time # Family model ids: nano-banana-pro, nano-banana-2, nano-banana-2-lite # 1. Submit Task headers = {"Authorization": "Bearer $API_KEY"} resp = requests.post( "https://api.apipod.ai/v1/images/generations", headers=headers, json={ "model": "nano-banana-2", "prompt": "A futuristic banana..." } ) task_id = resp.json()["data"]["task_id"] # 2. Poll Status while True: res = requests.get( f"https://api.apipod.ai/v1/images/status/{task_id}", headers=headers ).json() data = res["data"] if data["status"] == "completed": print(data["result"][0]) break time.sleep(1)
Frequently Asked Questions
Everything you need to know about the Nano Banana model family.
What is the Nano Banana family?
Nano Banana is Google DeepMind's Gemini family of image generation and editing models. APIPod offers four: Nano Banana Pro (Gemini 3 Pro Image) for studio-grade work; Nano Banana 2.1, the next generation with better realism and text at a lower price; Nano Banana 2 (Gemini 3.1 Flash Image) for Pro-level quality at Flash speed; and Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image), the fastest and most cost-effective.
Which Nano Banana model should I choose?
Pick Pro when text accuracy and top quality matter most (print, infographics, storyboards). For everyday production, start with Nano Banana 2.1 — better images and text from $0.02 per 1K image. Choose Nano Banana 2 if you need 1:4, 4:1, 1:8 or 8:1 banner ratios or 512px output, and Lite for real-time, high-volume work where 1K is enough.
How is the Nano Banana family priced?
All four are billed per generated image. Nano Banana Pro is $0.06 / $0.09 / $0.12 (1K / 2K / 4K); Nano Banana 2.1 is $0.02 / $0.04 / $0.05 (1K / 2K / 4K); Nano Banana 2 is $0.04 (512px / 1K), $0.06 (2K) and $0.08 (4K); Nano Banana 2 Lite is a flat $0.02 (1K). The console shows the live price before you submit.
How many reference images can I use?
Nano Banana Pro takes up to 8 reference images and keeps up to 5 people consistent. Nano Banana 2.1, 2 and 2 Lite take up to 14; 2.1 keeps up to 4 characters and 10 objects consistent across edit turns, and Nano Banana 2 keeps up to 5 characters and 14 objects. Pro and Nano Banana 2 also support Google Search grounding via the google_search parameter.
Can I use generated images commercially?
Yes. Images generated through the API come with commercial usage rights. All outputs include an invisible SynthID watermark for AI content identification, which does not affect visual quality.
Ready to create with Nano Banana?
Pick the model that fits your speed, quality and budget — all four run on the same unified API.