nano-banana-2-mcp

MCP server for Google Gemini image generation, upgraded for Nano Banana 2 (gemini-3.1-flash-image-preview).

A fork/rewrite of ConechoAI/Nano-Banana-MCP with:

• Nano Banana 2 model — better text rendering, thinking modes, resolution control
• Resolution control — 1K, 2K, 4K
• Aspect ratio — any ratio the API supports (1:1, 16:9, 9:16, 4:3, etc.)
• Thinking modes — minimal or high
• Multiple images — generate 1–4 variations per call
• File-path-only mode — no inline base64, fixes context window overflow in Claude Code
• Security hardening — path validation, file size caps, no plaintext API key storage

Setup

1. Get a Gemini API key

Get one from Google AI Studio.

2. Install

Via Claude Code plugin (recommended):

claude plugin add nano-banana-2-mcp

Or manually via npx — add to your Claude Code MCP settings:

{
  "mcpServers": {
    "nano-banana-2": {
      "command": "npx",
      "args": ["-y", "nano-banana-2-mcp"],
      "env": {
        "GEMINI_API_KEY": "your-api-key-here"
      }
    }
  }
}

Or from source for development:

git clone https://github.com/daveremy/nano-banana-2-mcp.git
cd nano-banana-2-mcp
npm install
npm run build

Then point your MCP config at dist/index.js:

{
  "mcpServers": {
    "nano-banana-2": {
      "command": "node",
      "args": ["/path/to/nano-banana-2-mcp/dist/index.js"],
      "env": {
        "GEMINI_API_KEY": "your-api-key-here"
      }
    }
  }
}

3. Restart Claude Code

The tools will be available after restart.

Tools

generate_image

Generate a new image from a text prompt.

Parameter	Type	Default	Description
prompt	string	(required)	Text prompt for the image
aspectRatio	string	"1:1"	Aspect ratio (e.g. "16:9", "9:16")
resolution	string	"1K"	1K, 2K, or 4K
thinking	string	"minimal"	minimal or high
numberOfImages	number	1	1–4
returnInlineImage	boolean	true	If false, return only file path

edit_image

Edit an existing image file.

Same parameters as generate_image, plus:

Parameter	Type	Description
imagePath	string	(required) Path to the image to edit
referenceImages	string[]	Optional additional reference images

continue_editing

Continue editing the last generated/edited image. Same parameters as edit_image minus imagePath (uses the last image automatically).

get_configuration_status

Check API key, active model, and settings.

get_last_image_info

Get path and size of the last generated image.

Environment Variables

Variable	Default	Description
GEMINI_API_KEY	(required)	Google Gemini API key
NANO_BANANA_MODEL	gemini-3.1-flash-image-preview	Model ID override
NANO_BANANA_OUTPUT_DIR	./generated_imgs	Image save directory
NANO_BANANA_INLINE_IMAGE	true	Default for returnInlineImage

Capability Gating

The server auto-detects model capabilities:

• Image models (*-image, *-image-preview): get imageConfig (resolution, aspect ratio)
• Gemini 3.x image models: also get thinkingConfig
• Other models: basic config only

This means you can use NANO_BANANA_MODEL=gemini-2.5-flash-image and it will send imageConfig but skip thinkingConfig.

Claude Code Plugin

This repo includes a Claude Code plugin with a generate-image skill that provides best-practice prompting guidance. Install via claude plugin add nano-banana-2-mcp or add the repo path to your Claude Code plugins config.

Contributing

See CONTRIBUTING.md for development setup, testing, and release process.

Attribution

Based on ConechoAI/Nano-Banana-MCP (MIT License).

License

MIT