Google DeepMind

Nano Banana 2 Preview

Generally availableProprietary66K contextReasoning

Summary

Nano Banana 2 Preview is a proprietary language model from Google DeepMind, released on 26 Feb 2026. It accepts images and text and generates text and images, with a 66K-token context window and up to 59K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.50 per million input tokens and $3 per million output tokens (OpenRouter), across 1 provider we track.

Image model for prompt-driven generation, editing, and visual design workflows

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperGoogle DeepMind
Release date26 Feb 2026 [models.dev]
StatusGenerally available [models.dev]
WeightsProprietary (API only) [models.dev]
Context window66K (65,536 tokens) [models.dev]
Max output59K tokens [models.dev]
Knowledge cutoffJan 2025 [models.dev]
InputImages and text [models.dev]
OutputText and images [models.dev]
Reasoning modeYes [models.dev]
Tool callingNo [models.dev]
Structured outputYes [models.dev]

Nano Banana 2 Preview API pricing by provider

USD per million tokens. Sorted by input price.

ProviderInputOutputCached inputContextSource
OpenRouter
google/gemini-3.1-flash-image-preview
$0.50$3–66KOpenRouter, 18 Nov 2026

Frequently asked questions

What is the context window of Nano Banana 2 Preview?

Nano Banana 2 Preview has a context window of 65,536 tokens (66K) and can generate up to 58,982 tokens in a single response, according to models.dev.

How much does the Nano Banana 2 Preview API cost?

Nano Banana 2 Preview is listed from $0.50 per million input tokens and $3 per million output tokens (OpenRouter). Prices last verified 18 Nov 2026.

Is Nano Banana 2 Preview open source?

No. Nano Banana 2 Preview is a proprietary model; Google DeepMind has not released its weights. It is available through APIs including OpenRouter.

When was Nano Banana 2 Preview released?

Google DeepMind released Nano Banana 2 Preview on 26 Feb 2026, according to models.dev.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
Qwen3-Omni FlashAlibaba (Qwen)15 Sept 202566K$0.43$1.66Closed
Aion-2.0Aion Labs21 Dec 2025131K$0.80$1.60Closed
Seed 2.0 CodeByteDance Seed14 Feb 2026256K$0.50$3Closed
Seed 2.0 ProByteDance Seed14 Feb 2026256K$0.50$3Closed
GLM-5-TurboZhipu AI (Z.ai)16 Mar 2026200K$1.20$4Closed
Aion-3.0-MiniAion Labs14 May 2026131K$0.70$1.40Closed

Change history

  • No longer listed on vercel
  • Name changed from Gemini 3.1 Flash Image Preview (Nano Banana 2) to Nano Banana 2 Preview
  • Input modalities changed from text, image to image, text
  • Max output changed from 33K to 59K
  • Context window changed from 131K to 66K
  • Name changed from Nano Banana 2 Preview to Gemini 3.1 Flash Image Preview (Nano Banana 2)
  • Context window changed from 66K to 131K
  • Max output changed from 59K to 33K
  • Input modalities changed from image, text to text, image
  • Name changed from Gemini 3.1 Flash Image Preview (Nano Banana 2) to Nano Banana 2 Preview
  • Context window changed from 131K to 66K
  • Input modalities changed from text, image to image, text
  • Structured output changed from none to yes
  • Max output changed from 33K to 59K
  • Now listed on OpenRouter
  • Now listed on vercel
  • Added to the directory

Sources