curl-X POST 'https://pollo.ai/api/platform/v1/generation/google/nano-banana/image'\-H'Content-Type: application/json'\-H'x-api-key: YOUR_API_KEY'\-d'{ "input": { "prompt": "A photorealistic portrait of an astronaut relaxing on a tropical beach, golden hour lighting, 85mm lens", "aspectRatio": "1:1" }}'
Optionally include webhookUrl to receive a callback when the task succeeds or fails.
Optional request field
{"input":{"prompt":"A photorealistic portrait of an astronaut relaxing on a tropical beach, golden hour lighting, 85mm lens","aspectRatio":"1:1"},"webhookUrl":"https://example.com/webhooks/pollo"}
Nano Banana is useful when image generation and image editing need to share the same prompt language. It can create new images, revise uploaded images, or combine multiple inputs into one output.
With Nano Banana on Pollo API, developers can add text-to-image, image-to-image, and multi-image-to-image workflows for creator tools, commerce systems, design apps, and visual automation.
Key Features of Nano Banana API
Natural Language Editing
Nano Banana accepts plain editing instructions, so users can ask for changes to objects, backgrounds, colors, lighting, style, or composition without masking tools.
Character Consistency
The model is suited to workflows where the same person, mascot, product, or object needs to remain recognizable across related image edits.
Multi-Image Fusion
Multiple input images can guide one output, making it useful for combining subjects, inserting objects, matching references, or building composite visuals.
Text and Image Guidance
Nano Banana can start from a prompt, a source image, or several reference images, depending on whether the task is generation, editing, or fusion.
Iterative Visual Refinement
Instructions can focus on one change at a time, helping users adjust a generated or uploaded image through smaller, easier-to-review steps.
Style Transfer and Restyling
The model can apply style changes while using the source image or references to preserve important subject, layout, or object details.
Use Cases of Nano Banana API
Character Asset Variants: Generate the same character in new outfits, poses, scenes, or moods while keeping key identity details recognizable.
Product Image Edits: Change backgrounds, props, lighting, or usage context around a product image without rebuilding the full scene manually.
Multi-Reference Composites: Combine a person, object, room, or style reference into one coherent image for design previews or campaign drafts.
Creator Editing Tools: Let users revise portraits, thumbnails, stickers, avatars, and social visuals with normal written instructions.
E-Commerce Scene Swaps: Place a catalog item into kitchens, offices, shelves, outdoor spaces, or lifestyle settings for quick visual testing.
Storyboard and Concept Art: Produce consistent scene stills, prop explorations, environment variations, or character sheets for early creative planning.
Localized Visual Variants: Adapt backgrounds, packaging context, seasonal details, or layout style while keeping the main subject stable.
Automated Image Workflows: Build batch systems that apply repeatable edits or reference combinations to large image sets.
How to Use Nano Banana API
Get an API Key: Create a Pollo API account and generate your API key from the developer dashboard.
Choose the Model: Call the Nano Banana route at /generation/google/nano-banana/image.
Add Your Inputs: Send a prompt and aspectRatio for text-to-image, or add images for image editing and multi-image workflows; set resolution when needed.
Generate and Retrieve: Submit the request, save the returned taskId, and retrieve the image when the task status is complete.
Prompting Best Practices for Nano Banana API
Write Nano Banana prompts like concise edit or composition instructions. Define the source-image role, target change, preserved details, visual style, and output use.
A Simple Prompt Formula
Source or subject + requested change + details to preserve + reference role + style or lighting + output format
What You Should Notice
Preserved Details: State what should not change, such as face shape, product logo, object geometry, clothing color, or camera angle.
Reference Roles: Explain what each image contributes, such as character identity, background, object insert, palette, pose, or material style.
Single-Step Edits: Ask for one main change per request when accuracy matters, then iterate with follow-up edits.
Composition Scope: Describe crop, orientation, spacing, and subject placement before adding smaller style details.
Material and Lighting: For product or room edits, specify surfaces, shadows, reflections, and light direction to reduce mismatched composites.
Text Caution: Keep any in-image text short and easy to verify, especially for labels, signs, or packaging mockups.
Example Prompts
Character Outfit Variant
"Use the uploaded character reference. Keep the same face, hair shape, glasses, and body proportions, but change the outfit to a green field jacket and hiking boots. Place the character beside a trail map kiosk in soft morning light."
Furniture Room Composite
"Combine the uploaded armchair and living room images. Place the chair near the window at a realistic scale, match the room's shadows and floor reflections, preserve the chair fabric texture, and remove any extra furniture that blocks it."
Menu Board Restyle
"Edit the cafe menu board photo into a cleaner chalkboard style. Keep the same item order and prices, simplify the decorative border, improve contrast, and make the board look photographed in the same wall position."
Packaging Context Edit
"Use the uploaded granola pouch image. Keep the pouch shape, label layout, and flavor name unchanged. Move it onto a kitchen counter with oats, dried fruit, and a ceramic bowl nearby, using natural window light."
Education Diagram Visual
"Create a square classroom visual showing how rainwater moves from roof gutters into a storage barrel and then to garden plants. Use simple labeled parts, clear arrows, friendly colors, and enough spacing for young students to understand."
Nano Banana vs Nano Banana 2 vs Nano Banana Pro
Capability
Nano Banana
Nano Banana 2
Nano Banana Pro
Model Position
Gemini 2.5 Flash Image-based workflow for generation and editing
Newer general-purpose Nano Banana image model
More complex visual production and instruction-heavy workflows
Main Workflow
Text-to-image, image editing, and multi-image fusion
Broader image generation and reference processing
Detailed composition, brand-sensitive assets, and complex prompts
Reference Handling
Useful for subject consistency and image-guided edits
Better fit when multiple references and consistency are central
Better fit when reference accuracy and layout control are strict
Iteration Style
Short, direct edits and compositing requests
General creative iteration with stronger output options
More deliberate instructions with tighter visual requirements
Output Planning
Works well with clear edits, stable subjects, and simple composites
Suitable for more varied image generation tasks
Suitable for controlled layouts, localized visuals, and precise asset direction
Recommended For
Creator tools, product edits, character variants, and reference composites
General-purpose image generation and editing apps
Brand, campaign, editorial, and production-facing visual systems
Why Choose Nano Banana API?
Nano Banana is a practical option when users need one model for generation, editing, and reference-based image composition. It is especially useful for prompt-driven visual tools.
Through Pollo API, developers can run Nano Banana alongside 300+ image and video models, with one API surface, clean documentation, scalable infrastructure, and human support.
Use Nano Banana on Pollo API when your product needs natural-language edits, subject consistency, multi-image fusion, or repeatable image automation.
Nano Banana API FAQs
What is Nano Banana?
Nano Banana is Google's Gemini 2.5 Flash Image model for generating and editing images from text, uploaded images, or multiple image references.
What inputs does Nano Banana API support on Pollo API?
Nano Banana API supports text-to-image, image-to-image, and multi-image-to-image workflows. Developers can use prompts, a source image, or multiple reference images.
Is Nano Banana API useful for character consistency?
Yes. It can help keep a character, product, or object recognizable across related image edits when prompts clearly define what details must remain unchanged.
Can Nano Banana combine multiple images?
Yes. Nano Banana supports multi-image workflows, so developers can build tools that combine subjects, objects, rooms, poses, or style references into one output.
Does Nano Banana API support video or audio inputs?
No. Nano Banana is an image generation and editing model. For video workflows, use a video model such as Veo through Pollo API.
How is Nano Banana different from Nano Banana Pro?
Nano Banana is better for direct generation, editing, and reference composites. Nano Banana Pro is better suited to more complex prompts, stricter layouts, and production-style assets.
