Try “Veo”, “voiceover”, “thumbnail” or “Descript” · Esc to close

Nano Banana 2 🧠 Model Freemium

Google's Gemini image models (Nano Banana 2, Pro and 2 Lite) for generation and editing up to 4K

Image models · Free plan, paid from $4.99/mo

9.0editor score
Visit Nano Banana 2 ↗
Maker
Google DeepMind
Latest release
2026-02
Weights
Closed
Licence
–
Apps offering it
30

Where you can use Nano Banana 2

Apps in this directory that let you generate with Nano Banana 2, from their own documentation.

See every app × model on the model map →

Good for

About Nano Banana 2

What it is

Nano Banana is Google DeepMind's name for its Gemini image generation and editing models. Nano Banana 2 (Gemini 3.1 Flash Image) launched on 26 February 2026 and is the default across Google's products. The family also includes Nano Banana Pro (Gemini 3 Pro Image, highest fidelity) and Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image, 30 June 2026, fastest and cheapest). A "Nano Banana 2.5 Flash" has only been spotted in code, not announced.

Key features

  • Outputs from 512px to 4K across multiple aspect ratios
  • Subject consistency for up to 5 characters and 14 objects
  • Precise text rendering and in-image translation
  • Real-time web search grounding for accurate landmarks, products and places

Where you can use it

In the Gemini app, Google Search (AI Mode, Lens), Ads, Photos and Google Flow (zero credits for Nano Banana 2), plus Google AI Studio, the Gemini API and Google Cloud. Third-party apps that document it: Adobe Firefly (Gemini 3.1 with Nano Banana 2), Krea, Picsart, Higgsfield (Nano Banana Pro), CapCut and Dreamina (Nano Banana Pro) and ComfyUI partner nodes.

Pricing and rights

Gemini API: Nano Banana 2 output is billed at $60 per million image tokens (about $0.067 per 1K image by third-party estimates); Nano Banana 2 Lite is about $0.034 per 1K image. The Gemini app includes image generation for free users with limits; Google AI Plus is $4.99 per month in the US and Pro $19.99 with higher Nano Banana limits. All outputs carry an invisible SynthID watermark, and Google pairs this with C2PA Content Credentials.

Who it is for

Social media creators, marketers and storytellers who need fast edits and consistent characters across many images.

Verdict

Nano Banana 2 combines speed, 4K output and strong consistency, and it is available free or cheaply in many apps. Versions and names change often, so confirm which variant an app runs, and expect Google's filters on edits of real people.

image-generation google gemini editing character-consistency

Pros

  • 512px to 4K output in many aspect ratios
  • Keeps up to 5 characters and 14 objects consistent across images
  • Uses web search for accurate real-world subjects
  • Free in the Gemini app and at zero credits in Flow; SynthID plus C2PA on outputs

Cons

  • Around $0.067 per 1K image on the API (third-party estimate), more than Imagen 4 was
  • Several versions with similar names (Nano Banana, 2, Pro, 2 Lite) make app menus confusing
  • The original Nano Banana (Gemini 2.5 Flash Image) shuts down on 2 October 2026
  • Closed model with Google content filters

Similar AI models

All image models →

GPT Image 2.5 🧠 ModelFreemium

OpenAI's image model behind ChatGPT Images, with 4K output and up to 16 reference images

OpenAI · 2026-09

9.0 Visit ↗

FLUX.2 🧠 ModelOpen source

Black Forest Labs' FLUX.2 family: hosted [max]/[pro]/[flex] plus open-weight [dev] and [klein]

Black Forest Labs · 2026-01 · open weights

8.6 Visit ↗

Midjourney V8.2 🧠 ModelPaid

Midjourney's own image model, known for aesthetics and personalization, now in V8.2

Midjourney · 2026-07

8.5 Visit ↗

Seedream 5.0 Pro 🧠 ModelFreemium

ByteDance's Seedream image models, used in CapCut, Dreamina and many creative apps

ByteDance Seed · 2026-07

8.3 Visit ↗

Ideogram 4.0 🧠 ModelOpen source

Ideogram's text-rendering image model with JSON prompting, now also as non-commercial open weights

Ideogram · 2026-06 · open weights

8.0 Visit ↗

Qwen-Image 3.0 🧠 ModelOpen source

Alibaba's Qwen image models, known for dense multilingual text rendering, with open-weight versions

Alibaba (Qwen team) · 2026-09 · open weights

More from Google DeepMind

Veo 3.1 🧠 ModelPaid

Google DeepMind's text- and image-to-video model with native sound, up to 4K

Google DeepMind · 2025-10

8.8 Visit ↗

Gemini 3.1 Pro 🧠 ModelFreemium

Google's Gemini models (3.1 Pro, 3.8 Flash) in the Gemini app, NotebookLM and the Gemini API

Google DeepMind · 2026-02

8.7 Visit ↗

Gemini 3.8 Flash TTS 🧠 ModelPaid

Google's prompt-directed TTS with 2,000+ voices, 100+ languages and consented cloning

Google DeepMind · 2026-09

8.0 Visit ↗

Lyria 3.5 🧠 ModelPaid

Google DeepMind's music model for full songs with vocals, lyrics and SynthID

Google DeepMind · 2026-07

8.0 Visit ↗

Popular searches