Nano Banana 2 Lite · Image Editing

ABOUT THE WORKFLOW

Edit an Image in Seconds
Upload the image you want to change and describe the edit in plain language. Nano Banana 2 Lite applies it and returns a 1K resolution result in about 4 seconds. Add up to 13 more reference images to bring in characters, products, styles, or objects from other sources. Leave the image inputs empty and write a full scene description to generate from scratch instead.

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 Lite (Gemini 3.1 Flash Lite Image) by Google. The fastest and most cost-efficient model in the Nano Banana family, built for high-throughput editing and generation at $0.034 per image with character consistency, accurate text rendering, and 14 supported aspect ratios.

HOW IT WORKS

Step 1. Upload your image
The image you want to edit. This goes into the first image slot. Add more images in slots 2 through 14 to bring in characters, styles, objects, or backgrounds from other sources.
Works great with: photos · product shots · portraits · AI-generated images · illustrations

Step 2. Write your prompt
Describe the change and what to keep. "Instead of red I want blue color and replace the axe with a black gun with blue glow" tells the model what to swap and what to preserve.

Step 3. Choose your aspect ratio and format
Pick from 14 aspect ratios (16:9, 9:16, 1:1, 4:3, and more). Output as PNG or JPEG.

Step 4. Hit run and download
Nano Banana 2 Lite processes the edit in about 4 seconds and returns the result at 1K resolution.
Ready for: Photoshop · Figma · Canva · social media · e-commerce · print

First time? Upload an image, write a short edit instruction, and hit run. Leave aspect ratio, format, and all other settings as-is.

RECOMMENDED SETTINGS

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

• Standard image edit — 16:9 aspect ratio, PNG, 1 image, thinking off. Upload your image, write the edit, run.

• Generate from scratch instead of editing — Leave all image inputs empty and write a full scene description in the prompt. The model generates a new image from text alone.

• Composite multiple references into one image — Upload separate reference images in slots 1 through 14. Describe how they combine: "Place the person from image 1 in the setting from image 2, wearing the outfit from image 3."

• Add text to the image — Put the exact words in quotes. "Add a neon sign reading 'OPEN DAILY' above the door." The model renders legible text directly in the output.

• Result quality not high enough for the edit — Turn the thinking level from "off" to a higher setting. The model reasons through complex edits more carefully with thinking enabled, at the cost of slightly longer generation.

• Edit changed too much or too little — Be more specific about what to change and what to keep. "Change only the jacket color from red to blue, keep everything else identical" narrows the scope.

Prompt: State what changes and what stays. "Replace the red color with blue and swap the axe for a glowing gun, keep the character pose and background" is specific. "Make it cooler" gives the model nothing to work with.

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

🛍️ High-Volume E-commerce Edits
Batch-edit product photos at scale: swap backgrounds, adjust staging, and generate color variants at $0.034 per image with 4-second turnaround.

📱 Social Media Asset Iteration
Generate dozens of ad variants, post layouts, and banner options from a single source image across multiple aspect ratios without waiting.

🎨 Multi-Reference Compositing
Combine characters, products, outfits, and backgrounds from up to 14 separate images into one composed scene with a single prompt.

🔤 Quick Mockups with Text
Generate product packaging mockups, signage concepts, or social graphics with legible text rendered directly into the image.

WHAT WORKS BEST / WHAT TO AVOID

✅ Works great

• Clear, specific edit instructions that state what changes and what stays

• Well-lit source images with defined subjects

• Multi-reference prompts that describe what each image contributes

• Text in quotes for accurate in-image rendering

⚠️ May produce softer results

• Vague instructions like "improve this" or "make it nicer"

• Very low-resolution or heavily compressed source images

• Conflicting edits in a single prompt

• Expecting 4K output (the model generates at 1K resolution)

FAQ

What is Nano Banana 2 Lite?
Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) is Google's fastest and most cost-efficient image model, released June 30, 2026. It generates and edits images in about 4 seconds at 1K resolution across 14 aspect ratios. It sits below Nano Banana 2 and Nano Banana Pro in Google's image model lineup, trading maximum creative control for speed and low cost.

How is Nano Banana 2 Lite different from Nano Banana Pro?
Nano Banana Pro (Gemini 3 Pro Image) is built for complex, professional use cases with advanced reasoning and supports up to 4K resolution. Nano Banana 2 Lite is built for speed and volume at 1K resolution. Use Lite for rapid iteration, batch generation, and high-throughput pipelines. Use Pro when accuracy and fine control matter more than speed.

How many reference images can I use at once?
Up to 14. Each image can carry a different element: a character, a product, a background, a style reference, an outfit, or an object. Describe what each reference contributes in the prompt so the model knows how to combine them.

Does Nano Banana 2 Lite render text inside images?
Yes. The model renders legible text directly into generated and edited images. Put the exact words in quotes in the prompt. This works for signage, labels, product packaging, and social media graphics.

What does the thinking level setting do?
Thinking level controls how much the model reasons before generating. With thinking off, the model responds fastest. Turning it on adds processing time but improves results for complex edits that require spatial reasoning, multi-step changes, or precise composition across multiple references.

Do the outputs have a watermark?
Google applies an invisible SynthID watermark to all images from Nano Banana 2 Lite. It does not change how the image looks, but it stays embedded in the file to mark it as AI-generated.

How to run Nano Banana 2 Lite image editing online?
You can run Nano Banana 2 Lite image editing online through Floyo. No installation, no setup, no API key to wire up. Open the workflow in your browser, upload your inputs, 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 runs an edit and likes the result. 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?
Upload your image, describe the edit, and hit run.

→ Launch Workflow, Free

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