Create and edit images with coherent reference blending, recognizable subjects, focused natural-language changes, and context-aware visual reasoning.
A flexible image model for combining, preserving, and changing what matters.
Reference blending
Bring subjects, objects, and environments from separate references into one coherent scene.
Subject consistency
Keep a recurring character or product recognizable across related visual directions.
Targeted edits
Describe one focused change while asking the model to preserve the rest of the image.
Context-aware scenes
Use Gemini's world knowledge to organize familiar objects and environments more meaningfully.
Key Features
Why creators still use the original Nano Banana.
The base Gemini 2.5 Flash Image model combines generation and editing in one natural-language workflow, with four strengths that are easy to apply to everyday visual work.
Feature 01 · Multi-image composition
Turn separate references into one believable scene.
Nano Banana can combine visual information from multiple image inputs, helping you place a subject, product, and environment into a new composition with more coherent lighting and perspective.
Combine complementary subject and environment references
Carry useful colors, materials, and shapes into a new scene
Describe how each reference should contribute to the result
Feature 02 · Character consistency
Keep the same subject recognizable across a story.
Reference-guided generation helps preserve identifying details when a recurring character or product moves into a new pose, activity, setting, or campaign frame.
Repeat distinctive clothing, colors, and accessories
Build related scenes around one recognizable subject
Create a more coherent visual sequence from a clear anchor
Feature 03 · Natural-language editing
Change one part without rebuilding the whole image.
Ask for a focused addition, removal, color change, or atmosphere shift, then state which subject, framing, materials, and surrounding details must remain unchanged.
Add, remove, recolor, or restyle a specific element
Preserve the composition and important subject details
Iterate with smaller instructions that are easier to review
Feature 04 · Gemini world knowledge
Give imaginative scenes a stronger sense of context.
Gemini's world knowledge can help the model reason about familiar places, objects, and relationships when composing educational, editorial, and concept imagery.
Organize familiar objects within a plausible environment
Turn broad topics into more specific visual concepts
Create a useful first draft that still receives a factual review
Examples + prompt recipes
See one visual world transformed in four different ways.
These examples follow the same fictional botanical explorer, teal field camera, and glass greenhouse so you can compare how Nano Banana blends references, preserves a subject, edits a scene, and builds context.
Blend a character, object, and location
Three complementary references become one naturally lit expedition campaign image.
View the prompt+
Use the uploaded explorer portrait, teal instant camera, and glass greenhouse as separate references. Create one cinematic 3:2 botanical expedition scene with the same explorer holding the teal camera inside the greenhouse. Match the warm morning light, realistic scale, perspective, reflections, and contact shadows. Preserve the mustard field jacket, coral scarf, camera shape, and greenhouse architecture. No text or logos.
The same clothing, accessories, and facial features connect four distinct field-journal moments.
View the prompt+
Using the uploaded explorer portrait as the identity anchor, create a four-frame editorial story sheet. Show the same explorer cataloguing leaves in a greenhouse, photographing an alpine flower, reviewing specimens at a wooden desk, and walking through light rain. Preserve facial features, short dark hair, mustard field jacket, coral scarf, teal instant camera, and realistic photographic style in every frame. No text or logos.
A focused edit moves the greenhouse from summer morning to early winter without changing the subject.
View the prompt+
Using the uploaded greenhouse expedition image, change only the season and atmosphere to an early winter afternoon with a light layer of snow outside the glass and cooler blue daylight. Keep the explorer's identity, mustard jacket, coral scarf, teal camera, pose, greenhouse geometry, plants, camera angle, crop, and object positions unchanged.
A broad topic becomes a detailed scene with recognizable research and plant-care zones.
View the prompt+
Create a detailed 3:2 editorial cutaway of a fictional alpine botanical field station. Include a glass propagation greenhouse, specimen photography table, seed storage cabinet, compact weather instruments, rainwater collection, drying racks, field notebooks, and a small mudroom for expedition gear. Make the spatial relationships plausible, use restrained natural materials and soft overcast mountain light, and include no labels, logos, or text.
Start from a written idea, or add clear references when the result needs to preserve a person, product, place, or visual direction.
01
Choose your starting point
Use Text to Image for a new scene, or upload references for an Image to Image workflow.
02
Describe the relationships
Explain what to combine or change, then state which identity, objects, framing, and style must remain.
03
Generate and review
Check the result closely, refine one instruction at a time, and download the strongest version.
Nano Banana questions
What is Nano Banana?+
Nano Banana is the widely used name for Gemini 2.5 Flash Image, Google's image generation and editing model. It accepts text and image inputs and can return image results.
What is the relationship between Nanobana, Nano Banana, and Google?+
Nano Banana is developed by Google. Nanobana is an independent creative platform that provides a web workflow for supported third-party models, including Nano Banana. Nanobana does not develop or own the model and is not affiliated with or endorsed by Google.
Do I need a Google or Gemini account to use Nano Banana on Nanobana?+
No separate Google or Gemini login is required for this workflow. Sign in to Nanobana and use the model through the Nanobana generator, subject to the credits and plan shown in your Nanobana account.
Which Nano Banana modes does Nanobana support?+
Nanobana supports Text to Image for creating from a written prompt and Image to Image for working from one or more uploaded references.
What makes a useful set of reference images?+
Choose sharp, complementary images with clearly visible subjects. Give each reference one clear role, such as identity, product shape, environment, or visual style, and explain those roles in the prompt.
How can I keep a character or product more consistent?+
Use a clear anchor image and repeat the distinguishing features that must remain, such as facial features, clothing, color, proportions, materials, and accessories. Change fewer variables in each generation.
How should I write a targeted edit?+
Describe the one element you want to add, remove, recolor, or restyle, then explicitly list the subject, framing, lighting relationships, and surrounding details that should stay unchanged.
Is the base Nano Banana model the newest version?+
No. Google currently describes Gemini 2.5 Flash Image as the legacy pioneer of the Nano Banana family. Nanobana lists newer Nano Banana models separately, so you can choose the version that best fits the current task.
What should I review before using a generated image?+
Check faces, hands, small objects, reflections, text, brand details, and factual or educational content. Generated images can miss instructions, so edit or regenerate anything that does not match the brief.
Make your first Nano Banana image.
Start with one of the prompt recipes above or describe your own scene. The generator is ready with the base Nano Banana model selected.
