Nano Banana 2 · Image Generation & Editing

ABOUT THE WORKFLOW

Generate or edit an image
Type a prompt to generate an image from scratch, or upload a reference and describe what to change. The model can also blend two references into one. It renders readable text and can pull real-world details from web search for accuracy.

Partner node. This workflow calls an external API, so each run uses credits from your API wallet. No API key needed. Floyo handles the connection.

Model

• Nano Banana 2 by Google DeepMind, the Flash-speed version of Gemini's image model (Gemini 3.1 Flash Image). It generates and edits images up to 4K, holds several characters and objects consistent across a scene, renders readable text, and can ground on live web search. Every output carries an invisible SynthID watermark.

HOW IT WORKS

Step 1. Write your prompt
Describe what to create, or how to change a reference image.
Works great with: photos · product shots · marketing assets · text and typography

Step 2. Add a reference image (optional)
Connect an image to edit it, and a second image to blend two references into one.

Step 3. Hit run and download
The model generates or edits the image in one pass and saves a PNG. Preview it, then download.
Ready for: Photoshop · Figma · Canva · any editor

First time? Leave every setting as-is. The defaults (1 image · auto ratio · 1K) are the right starting point for almost everyone.

RECOMMENDED SETTINGS

Quick-start guide. Find the goal that matches yours and copy the settings.

• Standard use (most people) — 1 image, auto aspect, 1K. The right starting point for almost everyone.

• Want higher resolution — raise the image size to 2K or 4K. It sharpens the result and costs more per image.

• Want options to choose from — raise the number of images to generate several in one run and keep the best.

• Change the shape — set the aspect ratio to match where the image is going, like 9:16 for stories or 16:9 for banners.

• Editing a reference — describe what stays the same and what changes, since the model follows specific instructions closely.

• Want the exact same result again — set a fixed seed in place of random. Random gives you a fresh take on each run.

Prompt: For a new image, describe the subject, setting, lighting, and style. For an edit, say what to keep and what to change, like "keep the pose and background, change the jacket to red leather." Spell out any text you want in the image, and name real landmarks or brands you want rendered accurately.

LEARN

📹 Videos

• Intro to Floyo

• ComfyUI 101 Free Course ft. Sebastian Kamph

• Floyo 101 for Team Collaboration

✨ Quick links

• Help docs

• LoRA Training guide

• Discover more workflows

USE CASES

🎨 Image Generation
Create photos, product shots, and marketing assets from a prompt.

✏️ Reference Editing
Change an outfit, background, or detail in a photo with a written instruction.

🧩 Multi-image Compositing
Blend two references, like a subject and a scene, into one image.

🔤 Text & Marketing Assets
Render readable text for posters, menus, and branded graphics.

WHAT WORKS BEST / WHAT TO AVOID

✅ Works great

• Specific prompts that name the change

• Reference images for edits and compositing

• Readable text and marketing layouts

• Real landmarks or brands to render accurately

⚠️ May produce softer results

• Small faces in a crowded frame

• Long passages of fine text

• Vague instructions like "make it better"

• Complex details that need a careful check

FAQ

What is Nano Banana 2?
Nano Banana 2 is Google DeepMind's Gemini 3.1 Flash Image model, released in February 2026. It brings the world knowledge, text rendering, and subject consistency of the Pro tier to a faster, cheaper Flash model, and it handles generation, editing, and multi-image compositing up to 4K. This workflow runs it as a single generate-and-edit tool.

How is Nano Banana 2 different from Nano Banana Pro?
Nano Banana 2 runs Pro-level features, like 4K output, readable text, and multi-character consistency, on the faster Flash stack at roughly half the cost. Nano Banana Pro, built on Gemini 3 Pro Image, is the premium tier for the highest factual accuracy and control. Nano Banana 2 also adds image-search grounding that Pro does not have, so use this one for speed and volume and Pro for maximum fidelity.

Can it edit images and blend multiple references?
Yes. Upload one reference image to edit it with a prompt, or connect a second reference to composite two images into one. For edits, describe what stays the same and what changes, and the model holds several characters and objects consistent across the scene.

Does Nano Banana 2 use web search?
Yes. The model can ground on Google Search, including image search, to render specific landmarks, brands, and current events accurately. That real-world grounding is one of the things that sets it apart from image models that rely only on pretraining.

Do the outputs have a watermark?
Yes. Every image from Nano Banana 2 carries an invisible SynthID watermark, paired with C2PA Content Credentials that record how the image was made. The watermark does not change how the image looks, but it stays embedded to mark the image as AI-generated.

Can I use Nano Banana 2 commercially?
Nano Banana 2 is a proprietary Google model with pay-per-use Flash-tier pricing, so commercial use is governed by Google's terms rather than an open license. Commercial use is generally supported, but review Google's current terms, note that outputs carry the SynthID watermark and C2PA credentials, and make sure you have the rights to any image you upload.

How to run Nano Banana 2 online?
You can run Nano Banana 2 online through Floyo. No installation, no setup, no API key to wire up. Open the workflow in your browser, type a prompt, and hit run. Free to try.

WHY FLOYO?

Floyo is the only platform with team collaboration for ComfyUI in the browser. You run workflows with no install. You share run history, assets, and models across your team. You pay only when you generate. Floyo supports open-source and closed-source models.

A designer generates an image and likes it. A teammate opens that exact run from shared history and keeps going. No file handoffs. No version confusion.

For studios and enterprise teams, Floyo adds private workspaces, pooled resources, and a team usage dashboard. Other ComfyUI cloud tools run for one person at a time. Floyo runs for the whole team, with transparent per-generation costs.

Ready to try it?
Type a prompt to generate, or add a reference to edit, then run it.

→ Launch Workflow, Free

Questions? Watch the free course or check the FAQ above.