Nano Banana 2.1

Google’s stable October 6 Gemini image API model, gemini-nano-banana-2.1, for image generation and conversational editing.

Image GenerationImage EditingGoogleMultimodal

Google Gemini API and AI Studio; no free API tier listed

Standard per 1M tokens: $1.50 input, $7.50 text/thinking output, $30 image output; batch half these rates. Listed square image-output equivalents: $0.0336/1K, $0.0504/2K, $0.0756/4K; other tokens and grounding can add cost.

Commercial

Visual examples

Google-published English illustrated photosynthesis recipe showing sunlight, water, carbon dioxide, a leaf kitchen, sugar and oxygen. — Google documentation example: an English photosynthesis illustration. Saved-image backend unverified; not a controlled model comparison.
Google AI for Developers · Open full-size image

🚀Function Overview

Generate or edit images at 1K, 2K and 4K with optional search grounding and reference images.

Key Features

131,072 input tokens and 32,768 output tokens
1K, 2K and 4K output; no legacy 512px output
Up to 14 references: four characters and ten objects
Web and image-search grounding; batch supported
Predecessor gemini-3.1-flash-image scheduled to shut down October 29, 2026

Use Cases

•Conversational image editing
•Reference-based character and object compositions
•Illustrated diagrams with human review

⚙️Input Parameters

array

generate_content contents: text and supported image/video/PDF parts.

config

object

SDK generation config: response_modalities and image_config with image_size/aspect_ratio.

💡Usage Examples

Example 1

Input Parameters

{
  "contents": [
    "Create an illustrated photosynthesis diagram with readable English labels."
  ],
  "config": {
    "response_modalities": [
      "TEXT",
      "IMAGE"
    ],
    "image_config": {
      "image_size": "1K"
    }
  }
}

Output Results

Unexecuted SDK keyword-argument example for client.models.generate_content(model="gemini-nano-banana-2.1", ...). The response contains candidates/content/parts; inspect text and inline image data.

Quick Actions

Technical Specifications

Hardware Type: Google Gemini API and AI Studio; no free API tier listed
Commercial Use: Supported
Pricing: Standard per 1M tokens: $1.50 input, $7.50 text/thinking output, $30 image output; batch half these rates. Listed square image-output equivalents: $0.0336/1K, $0.0504/2K, $0.0756/4K; other tokens and grounding can add cost.