Which integrations are available for this server?

Provides AI-powered image generation, editing, and combination using Google Gemini models (Nano Banana Pro, Nano Banana 2, Nano Banana). Supports text-to-image, image editing with reference images, and multi-image composition at various resolutions and aspect ratios.

How do I use mcp-nano-banana?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-nano-banana create a 1024x1024 image of a futuristic robot holding a banana" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

mcp-nano-banana

by codeFi

Overview Schema Related Servers Score Discussions

TypeScript

Hybrid

mcp-nano-banana

MCP server for AI-powered image generation using Google Gemini Nano Banana models. Generate, edit, and combine images from text prompts — directly from your AI coding assistant.

npm version license

Features

Text-to-image generation — Create images from natural language descriptions
Image editing — Upload images and modify them with text instructions
Multi-image combination — Mix up to 14 reference images into one composite
Resolution control — Generate at 512px, 1K, 2K, or 4K resolution
13+ aspect ratios — From square (1:1) to ultra-wide (21:9) and tall (1:8)
Auto-fallback — Pro model falls back to Flash on rate limits (429/403/402)
File output — Save to specific paths or return inline base64
Favicon generation — Generate complete favicon packages with one command
Multi-format output — PNG, JPEG, and WebP support

Related MCP server: Nano-Banana MCP Server

Quick Start

Prerequisites

Get a Gemini API key from Google AI Studio
Enable billing on your Google Cloud project — image generation requires a paid tier (~$0.04-0.05/image for NB2 at standard resolution)

Installation

npm install -g mcp-nano-banana

Set your API key:

export GEMINI_API_KEY=your-api-key

For MCP Registry publication metadata, the registry name is io.github.codeFi/mcp-nano-banana.

MCP Configuration

Codex

codex mcp add nano-banana --env GEMINI_API_KEY=your-api-key -- npx -y mcp-nano-banana@latest

Or add manually to ~/.codex/config.toml:

[mcp_servers.nano-banana]
command = "npx"
args = ["-y", "mcp-nano-banana@latest"]

[mcp_servers.nano-banana.env]
GEMINI_API_KEY = "your-api-key"

Qwen Code

Add to ~/.qwen/settings.json:

{
  "mcpServers": {
    "nano-banana": {
      "command": "npx",
      "args": ["-y", "mcp-nano-banana@latest"],
      "env": {
        "GEMINI_API_KEY": "your-api-key"
      }
    }
  }
}

Claude Code

claude mcp add nano-banana -e GEMINI_API_KEY=your-api-key -- npx -y mcp-nano-banana@latest

Or add manually to ~/.claude/claude_desktop_config.json:

{
  "mcpServers": {
    "nano-banana": {
      "command": "npx",
      "args": ["-y", "mcp-nano-banana@latest"],
      "env": {
        "GEMINI_API_KEY": "your-api-key"
      }
    }
  }
}

Other MCP Clients

For CLIs and desktop clients that use the standard mcpServers JSON format, register:

{
  "mcpServers": {
    "nano-banana": {
      "command": "npx",
      "args": ["-y", "mcp-nano-banana@latest"],
      "env": {
        "GEMINI_API_KEY": "your-api-key"
      }
    }
  }
}

Local / Development Build

git clone https://github.com/codeFi/mcp-nano-banana.git
cd mcp-nano-banana
npm install
npm run build

Then register with your MCP client:

{
  "mcpServers": {
    "nano-banana": {
      "command": "node",
      "args": ["/path/to/mcp-nano-banana/dist/index.js"],
      "env": {
        "GEMINI_API_KEY": "your-api-key"
      }
    }
  }
}

Models

Model	MCP Name	Description	Max Ref Images
`gemini-3-pro-image-preview`	Nano Banana Pro	High quality, thinking mode, professional assets	6 object + 5 character
`gemini-3.1-flash-image-preview`	Nano Banana 2	Fast, high-efficiency, 1:4/4:1/1:8/8:1 ratios	10 object + 4 character
`gemini-2.5-flash-image`	Nano Banana	Fastest, optimized for high-volume tasks	-

The server defaults to Nano Banana Pro for best quality and automatically falls back to Flash on rate limits.

Available Parameters

Resolution

Value	Approx. Pixels	Supported Models
`"512"`	~512px	Nano Banana 2 only
`"1K"`	~1024px	All models
`"2K"`	~2048px	All models
`"4K"`	~4096px	All models

Default: "1K"

Aspect Ratio

Value	Dimensions at 1K	Common Use Case
`1:1`	1024×1024	Social posts, icons
`16:9`	1344×768	Headers, presentations
`9:16`	768×1344	Stories, mobile
`4:3`	1152×896	Desktop, photos
`3:4`	896×1152	Portraits
`2:3`	~896×1344	Print, magazines
`3:2`	~1344×896	Photos
`4:5`	~896×1120	Instagram posts
`5:4`	~1120×896	Instagram landscape
`21:9`	~2048×878	Ultra-wide banners
`1:4`	~512×2048	Nano Banana 2 only — tall strips
`4:1`	~2048×512	Nano Banana 2 only — wide strips
`1:8`	~512×4096	Nano Banana 2 only — extra tall
`8:1`	~4096×512	Nano Banana 2 only — extra wide

Tools

`generate_image`

Generate or edit images from text prompts.

Parameter	Type	Required	Default	Description
`prompt`	string	✅ Yes	—	Text description or editing instructions
`imagePaths`	string[]	No	—	Reference image paths on disk (up to 14)
`model`	string	No	`gemini-3-pro-image-preview`	Model to use
`aspectRatio`	string	No	`1:1`	See table above for all options
`resolution`	string	No	`1K`	`512`, `1K`, `2K`, or `4K`
`outputFormat`	string	No	`png`	`png`, `jpeg`, `webp`
`outputPath`	string	No	—	Save image to this file path

Examples

Generate from text:

{
  "prompt": "A sunset over mountains with dramatic lighting",
  "aspectRatio": "16:9",
  "resolution": "2K",
  "outputPath": "/Users/currentUser/Desktop/sunset.png"
}

Edit an existing image:

{
  "prompt": "Change the background to a coffee shop interior",
  "imagePaths": ["/Users/currentUser/photo.jpg"],
  "outputPath": "/Users/currentUser/edited.png"
}

Combine multiple images:

{
  "prompt": "Put these glasses on this person's face",
  "imagePaths": ["/path/to/face.jpg", "/path/to/glasses.jpg"],
  "outputPath": "/path/to/result.png"
}

High-res wallpaper:

{
  "prompt": "A futuristic cityscape at night with neon lights and rain",
  "aspectRatio": "16:9",
  "resolution": "4K",
  "outputPath": "/path/to/wallpaper.png"
}

`generate_favicons`

Generate a complete favicon package from a prompt or existing image.

Parameter	Type	Required	Default	Description
`prompt`	string	No*	—	Icon description (required if `imagePath` not set)
`imagePath`	string	No*	—	Path to existing master icon image
`model`	string	No	`gemini-3-pro-image-preview`	Model for prompt generation
`outputDir`	string	No	—	Directory for individual favicon files
`outputPath`	string	No	—	Save ZIP package to this path
`returnZipBase64`	boolean	No	`false`	Return ZIP as base64 text
`appName`	string	No	`My App`	Manifest `name`
`shortName`	string	No	`App`	Manifest `short_name`
`themeColor`	string	No	`#ffffff`	Manifest theme color
`backgroundColor`	string	No	`#ffffff`	Manifest background color
`includeReadme`	boolean	No	`true`	Include README.md in package

Example

{
  "prompt": "A minimalist banana icon with a soft gradient",
  "outputDir": "/path/to/favicon",
  "outputPath": "/path/to/favicon.zip",
  "appName": "MyBrand",
  "shortName": "Brand",
  "themeColor": "#1a1a2e",
  "backgroundColor": "#ffffff"
}

Generated files:

favicon.ico — Multi-size icon (16/32/48px)
favicon-16x16.png — Browser tabs
favicon-32x32.png — Higher-res tabs
apple-touch-icon.png — iOS home screen
android-chrome-192x192.png — Android
android-chrome-512x512.png — Android HD
site.webmanifest — PWA manifest

HTML snippet (included in response):

<link rel="icon" href="/favicon.ico" sizes="any">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">

Pricing

Image generation requires a paid Google AI Studio account. Approximate costs:

Model	1K	2K	4K
Nano Banana 2	~$0.04	~$0.08	~$0.15
Nano Banana Pro	~$0.06	~$0.13	~$0.24

You only pay for what you generate — no monthly minimum.

Supported Image Formats

Input (reference images): png, jpg, jpeg, webp, gif Output: png, jpeg, webp

Development

# Build TypeScript
npm run build

# Watch mode
npm run dev

# Run tests
npm test

# Run the server directly
node dist/index.js

Troubleshooting

Issue	Solution
`GEMINI_API_KEY is not set`	Export the env var or add it to your MCP config
`429 Too Many Requests`	Server auto-falls back to Flash model. Add billing to your account.
`403/402 Quota exceeded`	Enable billing at Google Cloud Console
`503 Unavailable`	High demand — retry in a moment. Common with 4K requests.
Image not updating after edit	Make sure you rebuilt the TypeScript (`npm run build`)
`imagePaths` not found	Use absolute paths. Relative paths may not resolve correctly.

Architecture

AI Assistant (MCP Client)
       │  (MCP Protocol over stdio)
       ▼
mcp-nano-banana Server
       │  (HTTPS POST)
       ▼
Google Gemini API
       │
       ▼
Generated Image → saved to outputPath (or returned inline base64)

License

MIT

Author

codeFi

Available Tools

2 tools

generate_faviconsA

Generate a full favicon package from a prompt or an existing image. Creates PNG sizes, a web manifest, and optional README, then writes files and/or a ZIP bundle.

ParametersJSON Schema

Name	Required	Description	Default
`model`	No	Model to use when generating the master icon from a prompt.	gemini-3-pro-image-preview
`prompt`	No	Text description for generating a master icon (square). Required if imagePath is not provided.
`appName`	No	Manifest app name. Default: My App.
`imagePath`	No	Path to an existing master icon image. If provided, prompt is optional.
`outputDir`	No	Directory to write individual favicon files.
`shortName`	No	Manifest short name. Default: App.
`outputPath`	No	File path to save the favicon ZIP package.
`themeColor`	No	Manifest theme color. Default: #ffffff.
`includeReadme`	No	Include README.md in the package.
`backgroundColor`	No	Manifest background color. Default: #ffffff.
`returnZipBase64`	No	When true and outputPath is not provided, return the ZIP as base64 text.

TDQS

A3.9/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It discloses that files and/or a ZIP bundle are written and mentions optional README, but does not cover return behavior (returnZipBase64), overwrite semantics, or behavior when output destinations are missing. This is partial disclosure.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, front-loaded with the main action and outputs. No filler, every clause adds useful signal. This is highly concise and well-structured.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool has 11 parameters and no output schema, and while the description captures the core workflow, it leaves gaps about the interplay between outputDir, outputPath, and returnZipBase64. It also doesn't state what happens if no output destination is given, making it incomplete for complex usage.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema covers all 11 parameters with descriptions, so the baseline is 3. The description adds a high-level note about prompt vs imagePath and output formats, but doesn't enrich individual parameter semantics beyond what the schema already provides.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb+resource ('Generate a full favicon package') and enumerates concrete outputs (PNG sizes, web manifest, optional README, files/ZIP bundle). This clearly distinguishes it from the sibling generate_image, which is for generic image generation.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implicitly defines the tool's purpose by contrasting with generate_image and stating the input modes (prompt or existing image). However, it does not explicitly name alternatives or state when not to use this tool, so it lacks a full 'when-to-use' guide.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

generate_imageA

Generate an image using Google Nano Banana (Gemini) AI model. Can generate from text only, or use one or more reference images (up to 14) to guide generation.

ParametersJSON Schema

Name	Required	Description	Default
`model`	No	Model to use: gemini-2.5-flash-image (fast) or gemini-3-pro-image-preview (high quality, default)	gemini-3-pro-image-preview
`prompt`	Yes	Text description of the image to generate or instructions for editing/combining reference images
`imagePaths`	No	Optional array of file paths to reference images (up to 14). Supports png, jpg, jpeg, webp, gif.
`outputPath`	No	File path to save the generated image. If not provided, image data is returned inline.
`resolution`	No	Resolution of the generated image. 512 only on gemini-3.1-flash-image-preview. Nano Banana Pro supports 1K, 2K, 4K.	1K
`aspectRatio`	No	Aspect ratio of the generated image. Additional ratios (1:4, 4:1, 1:8, 8:1) available on gemini-3.1-flash-image-preview.	1:1
`outputFormat`	No	Output image format	png

TDQS

A4/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden of behavioral disclosure. It discloses the AI model and the capability to accept up to 14 reference images, but doesn't address return behavior or potential pitfalls. The outputPath schema covers the return mechanism, so partial credit is warranted.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, front-loaded with the core action and resource. Every word contributes to understanding the tool's purpose and main capabilities.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The description provides a high-level overview of the tool's capability, and the rich schema covers all parameters. Since there is no output schema, the description could mention the inline return behavior, but the outputPath parameter description already covers it. Adequate for a generation tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents all parameters. The description only repeats the 'up to 14' reference image limit, which is already in the imagePaths parameter description, adding no new semantic value.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's primary function: generating an image using the Google Nano Banana (Gemini) AI model. It distinguishes from the sibling tool generate_favicons by covering general image generation with text or reference images.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly identifies two usage modes: text-only generation and reference-image-guided generation. It does not mention alternatives or exclusions, but the context is clear enough for typical use cases.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

2 tool updatesv1.2.2Aug 7, 2026
- First observedgenerate_favicons
- First observedgenerate_image

Install Server

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.1/5.0

Disambiguation5/5

The two tools are clearly distinct: generate_image produces a single image from text or references, while generate_favicons creates a package of favicon files. There is no overlap in their purposes, so an agent can easily choose between them.

Naming Consistency5/5

Both tool names follow the same verb_noun pattern (generate_image, generate_favicons). The naming is perfectly consistent and predictable.

Tool Count4/5

With only 2 tools, the set is on the smaller side, but it is appropriately scoped for a server dedicated to Nano Banana image generation. The two tools cover both general image generation and a specific use case (favicons), so the count feels intentional rather than insufficient.

Completeness4/5

The tool surface covers the core generation capabilities: text-to-image and image-to-image via references, plus a specialized favicon output. Minor gaps exist (e.g., no explicit editing/upscaling tool, no model listing), but the server's stated purpose is well-served.

Maintenance

ActivityStale

ResponsivenessSyncing

Resources

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

mcp-nanobanana-proOAuth
io.github.AceDataCloud
MCP server for NanoBanana AI image generation and editing
Image & Video Processing Multimedia Processing
27
mcp-qwen-imageOAuth
io.github.AceDataCloud
MCP server for Qwen Image 3 AI image generation
AI & Machine Learning Image & Video Processing
1
mcp-flux-proOAuth
io.github.AceDataCloud
MCP server for Flux AI image generation
Image & Video Processing Multimedia Processing
23
mcp-veoOAuth
io.github.AceDataCloud
MCP server for Google Veo AI video generation
Multimedia Processing Image & Video Processing Entertainment & Media
23

Related MCP Servers

Nano Banana MCP Server
Image & Video Processing Multimedia Processing
NH-5
F
license
B
quality
D
maintenance
An MCP server that enables image generation using Google's Gemini Nano Banana Pro model via the Google AI Studio API. Users can generate and save images locally by providing text prompts through MCP-compatible clients.
Updated 7 months ago (2026-02-20 02:42 UTC)
1
-
Nano-Banana MCP Server
AI & Machine Learning Image & Video Processing
bongartzdiaz
A
license
A
quality
D
maintenance
An MCP server that provides AI image generation and editing capabilities using Google's Gemini 2.5 Flash Image API. It allows users to create new images from text, modify existing files, and perform iterative edits through natural language prompts.
Updated 5 months ago (2026-03-28 12:32 UTC)
6
758
MIT
MCP Banana Image
Image & Video Processing AI & Machine Learning Multimedia Processing
Trigi-Digital
A
license
Not graded
quality
C
maintenance
AI-powered image generation MCP server with 16 specialized tools for generating, editing, analyzing, and processing images using Google's Nano Banana 2 model. Supports custom API endpoints and integrates with AI coding assistants via natural language.
Updated 6 months ago (2026-03-01 02:55 UTC)
82
1
MIT
nano-banana-claude
AI & Machine Learning Image & Video Processing
tougenrip
A
license
Not graded
quality
D
maintenance
An MCP server that provides image generation using Google's Nano Banana Gemini models, with additional tools for background removal, upscaling, and format conversion via deterministic post-processing.
Updated 3 months ago (2026-06-09 18:15 UTC)
1
MIT

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/codeFi/mcp-nano-banana'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

mcp-nano-banana

Features

Quick Start

Prerequisites

Installation

MCP Configuration

Codex

Qwen Code

Claude Code

Other MCP Clients

Local / Development Build

Models

Available Parameters

Resolution

Aspect Ratio

Tools

generate_image

Examples

generate_favicons

Example

Pricing

Supported Image Formats

Development

Troubleshooting

Architecture

License

Author

Available Tools

TDQS

TDQS

Tool Schema Changelog

TDQS

Maintenance

Resources

Looking for Admin?

Related MCP Connectors

Related MCP Servers

Nano Banana MCP Server

Nano-Banana MCP Server

MCP Banana Image

nano-banana-claude

Latest Blog Posts

MCP directory API

`generate_image`

`generate_favicons`