aiaimodel.cards

Nano Banana 2

Image modelPrevious generationCard summarized

Powerful, high-efficiency image generation and editing, optimized for speed and high-volume use cases.

Source: Gemini 3.1 Flash Image Model Card (7 pp)

Use it via API

$0.067 / image

Where to call it

  • Gemini APIgemini-3.1-flash-image
  • Google Vertex AI

Supports

  • Reasoning control
  • Web search
  • Batch API
  • Image input
  • PDF input
Full feature support

Good for

  • Creation and editing with professional precision and quick iterations
  • Clear text for posters and intricate diagrams
  • Long-context real-world knowledge

Overview

Google · 04Nano Banana, front

Nano Banana 2

Powerful, high-efficiency image generation and editing, optimized for speed and high-volume use cases.

Typical price
$0.067 / image
1K (e.g. 1024×1024)
Max resolution
4K (4096×4096; up to 12288×1536 at 8:1)
Aspect ratios
14 options
Modes
text-to-image, image editing
Released
26 Feb 2026

Gemini 3.1 Flash Image (Nano Banana 2) is based on Gemini 3 Flash and generates images and text from text, image, audio and video input. In Google's human-rated Elo evals it leads Nano Banana, GPT-Image 1.5, Seedream 5.0 Lite and Grok Imagine Image Pro on every text-to-image and editing task shown, and matches or trails Nano Banana Pro on some editing tasks. p.2, p.5

gemini-3.1-flash-image
Typical price
$0.067 / image
Max resolution
4K (4096×4096; up to 12288×1536 at 8:1)
Aspect ratios
1:1, 1:4, 1:8, 2:3, 3:2, 3:4, 4:1, 4:3, 4:5, 5:4, 8:1, 9:16, 16:9, 21:9
Languages
Best in EN, ar-EG, de-DE, es-MX, fr-FR, hi-IN, id-ID, it-IT, ja-JP, ko-KR, pt-BR, ru-RU, ua-UA, vi-VN, zh-CN
Weights
Closed
Released
26 Feb 2026
  • text-to-image
  • image editing
  • multi-turn editing
  • multi-image fusion (up to 14 references)
  • Google Search grounding
  • Image Search grounding
  • video-to-image
  • interleaved text and images