Nano Banana 2
Nano Banana 2 is Google's fast AI image generator and editing model, with reliable text rendering and subject consistency
Nano Banana 2
Google’s fast Gemini image model for generating and editing photos from text, available on mitte.ai.
Nano Banana 2 is Google’s latest AI image model, also known as Gemini 3.1 Flash Image. It pairs the quality of Nano Banana Pro with the speed of Gemini Flash, so you can generate and edit images from a text prompt in seconds, with strong subject consistency, precise instruction following and reliable text rendering. On mitte.ai you can run Nano Banana 2 directly in your browser, or start from a ready-made preset.
Overview
| Model | Nano Banana 2 |
| Developer | Google (Gemini) |
| Type | Text-to-image and AI image editing |
| Resolution | 512px, 1K, 2K and 4K |
| Strengths | Speed, subject consistency, text rendering, instruction following |
What is Nano Banana 2?
Nano Banana 2 is an AI image model built on Google’s Gemini. You describe an image, or upload one and say what to change, and it renders the result in seconds. Because it draws on Gemini’s real-world knowledge, it renders specific subjects more accurately, and it is a clear step up on text inside images, which is where many image models struggle. It handles both creation from scratch and detailed edits, which makes it a strong fit for portraits, product shots, thumbnails, diagrams and design work.
Key features
- Fast generation and editing: Gemini Flash speed for rapid, iterative results.
- Reliable text rendering: readable, accurate text in images, a big upgrade over earlier Flash models.
- Precise edits: change the background, replace or add an object, shift day to night, change the camera angle or adjust focus, all from a sentence.
- Subject consistency: keep a face, character or product looking the same across edits.
- Real-world knowledge: pulls on Gemini to render specific subjects more accurately.
- Flexible output: resolution tiers from 512px up to 4K, across many aspect ratios.
How to use Nano Banana 2 on mitte.ai
- Open Nano Banana 2 on mitte.ai, or pick a preset below.
- Describe your image, or upload a photo and say what to change.
- Set the format (aspect ratio and resolution) for your use case.
- Generate. Nano Banana 2 renders the image in your browser.
- Refine. Keep editing by describing the next change, no need to start over.
Want to compare it with other image models first? Browse the image model collection.
Some Presets built with Nano Banana 2
The fastest way to use Nano Banana 2 is through a preset, each one tuned for a specific result and ready to run:
- Professional Headshot — polished headshots from a selfie.
- Heroic Portrait - Turn your photos to cinematic portrait photography
- Renaissance Portrait - Transform any photo into a classical oil painting.
- Miniature Portrait - Turn your photos to miniature portraits
- Face Swap - Instantly swap face in photos
(Confirm each preset slug against the live URL before publishing.)
FAQ
What is Nano Banana 2? Google’s latest Gemini image model (Gemini 3.1 Flash Image). It generates and edits images from text, fast, with strong subject consistency and reliable text rendering.
Who made Nano Banana 2? Google, as part of Gemini. On mitte.ai you can use it without any setup, directly in the browser.
How is it different from Nano Banana Pro? Nano Banana 2 brings Flash-level speed while keeping much of Pro’s quality, so it is built for fast, iterative generation and editing.
How do I use Nano Banana 2? Open it on mitte.ai or start from a preset above, describe your image (or upload one to edit), pick your format and generate.
What can Nano Banana 2 do? Create images from text and edit existing photos: change backgrounds, swap or add objects, adjust lighting and camera angle, render accurate text, and keep subjects consistent across edits.
What resolutions does it support? 512px, 1K, 2K and 4K, across a wide range of aspect ratios.