EvoLink.AIEvoLink.AI

Nano Banana 2 API

Nano Banana 2 API on EvoLink routes Gemini 3 Pro Image editing for precise object swaps, text fixes, and scene restyling from simple prompts. It is built for teams that need reliable, repeatable edits across batches, channels, and languages.

A highly dynamic, futuristic infographic poster for "Evolink," blending comic book art with high-end sci-fi concept art. A powerful, glowing cybernetic entity, the "Evo-Agent," bursts forth from a central data vortex platform. Data streams and light trails connect various floating holographic interfaces around the hero, forming a loose six-panel structure without rigid borders.Top section shows a sprawling, impossibly detailed digital landscape (8K Reality) transitioning into a massive holographic brain structure (Quantum Core). Mid-section features the Agent interacting with flowing text that translates instantly (Universal Decryption), flanked by complex floating holographic UI controls with sliders for manipulating visual parameters (Advanced UI). Bottom section displays floating screens in varied aspect ratios (1:1, 9:16, wide) showing identical content, next to multiple translucent echoes of the Agent in perfect pose consistency.The style is sleek, dominant tech-noir, with glowing circuit patterns and lens flares. Footer area with stylized minimalist node icons. Bold glowing tagline across the bottom: "EVOLINK: THE ARCHITECTURE OF FUTURE THOUGHT."

Parameters
size
auto
quality
2K
Estimated Cost
8.7 Credits
Sample Result
Sample generated image
1,170 (suggested: 2,000)

Upload up to 10 images (max 20MB each)

Click to upload or drag and drop

Supported formats: JPEG, JPG, PNG, WEBP
Maximum file size: 20MB; Maximum files: 10

Generated image 1

History

Max 20 items

0 running · 0 completed

Your generation history will appear here

Nano Banana 2 API for instruction-based image editing

Use the Nano Banana 2 API to edit photos, replace text, and keep character identity without manual masks. Build product, marketing, and app workflows with consistent outputs and high-resolution delivery.

Hero showcase of AI image editing feature 1

Capabilities of the Nano Banana 2 API

Semantic edits from plain language

The Nano Banana 2 API understands intent like 'remove the glare' or 'swap the mug for a glass' and applies the change while keeping the rest of the scene intact. This makes ecommerce cleanup, content refreshes, and social variations fast because you edit by describing outcomes, not by drawing masks.

Start using API
Workflow showcase of AI image editing feature 2

Clear text replacement and localization

Nano Banana 2 API is designed for clean text inside images, so posters, menus, price cards, and UI mockups stay readable after edits. It can replace wording, adjust labels, and keep layout balance, which helps teams localize assets across languages without rebuilding every design at scale.

Start using API
Workflow showcase of AI image editing feature 3

Multi-image fusion for consistent identity

For character or product consistency, the Nano Banana 2 API accepts up to 14 reference images in a single request. You can keep a model, mascot, or product line visually aligned across a series while changing scenes, angles, or backgrounds, which is ideal for catalogs and ads at scale.

Start using API
Workflow showcase of AI image editing feature 4

Why developers choose Nano Banana 2 API

Built on Gemini 3 Pro Image preview, Nano Banana 2 API delivers reliable edits, clear text rendering, and multi-image consistency, with official and Lite routes to balance stability and cost.

Instruction-first reliability

We follow precise object, text, and background instructions so batch edits stay predictable for product and marketing pipelines.

Production-grade resolution

Generate 1K/2K/4K outputs with legible text for posters, menus, diagrams, and UI mockups without extra cleanup.

Consistency plus cost control

Fuse up to 14 references to lock identity, then choose official or Lite routing to match budget and stability needs.

How to integrate the Nano Banana 2 API

Send an image, describe the change, and receive a ready-to-use result.

1

Step 1 — Upload or link images

Send a base image by URL or Base64, and include optional reference images for consistency.

2

Step 2 — Write the instruction

Describe the edit in plain language so the Nano Banana 2 API knows what to change.

3

Step 3 — Review and deploy

Receive the edited image, save it to your storage, and publish across channels.

Open API Playground

Key features of the Nano Banana 2 API

Built for editing-heavy production workflows

Control

Instruction-based semantic editing

The Nano Banana 2 API turns short prompts into targeted edits such as remove, replace, or restyle while preserving composition, so teams avoid manual masking and keep iteration fast.

Text

Clear text inside images

Gemini 3 Pro Image is built to render readable text for posters, menus, diagrams, and UI mockups, helping Nano Banana 2 API outputs stay legible across languages and formats.

Consistency

Multi-image fusion (up to 14)

Blend multiple references in one Nano Banana 2 API request to keep people, products, and visual style consistent across a series of edits or variations.

Look

Lighting and color adjustments

Refine mood by changing color grades, lighting direction, or day and night settings. This is useful for brand style alignment and seasonal campaigns.

Resolution

1K to 4K output options

Generate 1K, 2K, or 4K results and upscale for delivery, keeping edges crisp for web, apps, and print-ready assets.

Cost

Official and Lite routing

Choose the official Gemini 3 Pro Image preview route for stability, or use the Nano Banana 2 Lite route for lower cost when you can trade some control.

Nano Banana 2 API FAQs

Everything you need to know about the product and billing.

Nano Banana 2 API is our routing for the Gemini 3 Pro Image preview model (model name gemini-3-pro-image-preview). It focuses on instruction-based edits, text clarity, and multi-image consistency for production workflows. We also provide a Nano Banana 2 Lite route for lower-cost tests with different tradeoffs. This makes the Nano Banana 2 API a fit for product photos, marketing assets, and app content where repeatability matters and visual wording must stay readable.
Yes. The Nano Banana 2 API uses a single image generation endpoint that supports text-to-image, image-to-image, and image editing in one flow. Requests are processed asynchronously, returning a task ID you can poll for status. The response includes image links that are time-limited, so production workflows should save outputs to your own storage after completion. This unified flow keeps generation and editing consistent across the same model.
Nano Banana 2 API supports multi-image fusion with up to 14 reference images in a single request. This helps you keep a consistent person, mascot, or product line while changing scene context, camera angle, or background. For larger libraries, split requests into batches, then reuse the best outputs as fresh references. This approach keeps the visual identity stable across catalogs, ad variants, and seasonal updates.
Nano Banana 2 API is tuned for clear text inside images, which is valuable for posters, diagrams, menus, and UI mockups. It also supports bilingual text modification, so you can replace wording without breaking layout or typography. For best results, keep text blocks short, specify the desired language, and provide a clean reference image when editing existing signage. This balances readability with visual consistency.
Nano Banana 2 API delivers 2K native outputs with 4K upscaling available. The route also supports 1K, 2K, and 4K generation options. For most web and app use cases, start at 1K or 2K to keep iteration fast, then upscale to 4K for final hero images or print-ready assets. This workflow preserves detail while controlling cost.
We offer tiered pricing by route: the official gemini-3-pro-image-preview route is priced higher, while the Nano Banana 2 Lite route is lower cost. Current listed prices range around $0.107 to $0.122 per image for the official route and around $0.030 to $0.044 per image for Lite, with VIP and non-VIP tiers. Pricing can change, so confirm your current rates in the dashboard before scaling.
The official gemini-3-pro-image-preview route provides higher stability and controllability for demanding business scenarios. The Nano Banana 2 Lite route is more cost-effective, but it currently lacks size controls and can be slightly less stable. If you are launching production workflows, use the official route for consistency. If you are prototyping or running large-scale tests, Lite can reduce cost while you iterate.
Nano Banana 2 API is optimized for precise text alignment and instruction-following edits, which is valuable for text-heavy assets and brand-critical updates. Other image-edit APIs may excel in artistic looks or speed, so the best choice is to benchmark a few models with your own assets and evaluate which one follows instructions most consistently.
POST
/v1/images/generations

Generate Image

Create an image generation task using text prompts or reference images. Supports text-to-image, image-to-image, and image editing modes.

Asynchronous processing mode, use the returned task ID to .

Generated image links are valid for 24 hours, please save them promptly.

Request Parameters

modelstringRequiredDefault: gemini-3-pro-image-preview

Image generation model name.

ValueDescription
gemini-3-pro-image-previewNano Banana 2 model
Examplegemini-3-pro-image-preview
promptstringRequired

Prompt describing the image to be generated or how to edit the input image.

Notes
  • Limited to 2000 tokens
ExampleA cat playing in the grass
sizestringOptionalDefault: auto

Aspect ratio of the generated image.

ValueDescription
autoAuto detect
1:1Square (1:1)
2:3Portrait (2:3)
3:2Landscape (3:2)
3:4Portrait (3:4)
4:3Landscape (4:3)
4:5Portrait (4:5)
5:4Landscape (5:4)
9:16Vertical (9:16)
16:9Horizontal (16:9)
21:9Ultra-wide (21:9)
Example'16:9'
qualitystringOptionalDefault: 2K

Quality of the generated image.

ValueDescription
1KStandard quality
2KHigh quality (default)
4KUltra quality (Higher Cost)
Notes
  • 4K quality incurs additional charges
Example2K
image_urlsarrayOptional

Reference image URL list for image-to-image and image editing functions.

Notes
  • Max 10 images per request
  • Max size: 10MB per image
  • Formats: .jpeg, .jpg, .png, .webp
  • URLs must be directly accessible by the server
  • Maximum 5 real person images
Examplehttps://example.com/image1.png
callback_urlstringOptional

HTTPS callback address after task completion.

Notes
  • Triggered on completion, failure, or cancellation
  • Sent after billing confirmation
  • HTTPS only, no internal IPs
  • Max length: 2048 chars
  • Timeout: 10s, Max 3 retries
Examplehttps://your-domain.com/webhooks/image-task-completed

Request Example

{
  "model": "gemini-3-pro-image-preview",
  "prompt": "A cat playing on the grass",
  "size": "16:9",
  "quality": "2K"
}

Response Example

{
  "created": 1757165031,
  "id": "task-unified-1757165031-uyujaw3d",
  "model": "gemini-3-pro-image-preview",
  "object": "image.generation.task",
  "progress": 0,
  "status": "pending",
  "task_info": {
    "can_cancel": true,
    "estimated_time": 45
  },
  "type": "image",
  "usage": {
    "billing_rule": "per_call",
    "credits_reserved": 1.6,
    "user_group": "default"
  }
}