Skip to main content

MCP server for AI-powered image generation using Google Gemini models

Project description

gemini-image-mcp

MCP server for AI-powered image generation using Google Gemini models. Free API key, dual-model support, aspect ratio control.

Part of Channel 47, the open-source ecosystem of profession plugins for Claude Code. Get the newsletter for weekly skill breakdowns from production use.

Also known as "Nano Banana" — the friendly name for this image generation server.

Features

  • Dual Model Support: Flash (2-3s, 1024px) and Pro (4K quality)
  • Smart Model Selection: Auto-selects based on prompt keywords
  • Aspect Ratio Control: 1:1, 16:9, 9:16, 21:9, 4:3, 3:4, 2:1
  • Google Search Grounding: Factually accurate images (Pro only)
  • File Management: Upload, list, and delete files via Gemini Files API
  • Reproducible Generation: Seed support for consistent results

Installation

Via uvx (Recommended)

No installation needed - runs directly:

uvx gemini-image-mcp

Via pip

pip install gemini-image-mcp

Via pipx

pipx install gemini-image-mcp

Configuration

Environment Variable

Get your free API key from Google AI Studio.

export GEMINI_API_KEY="your-api-key-here"

Claude Code Integration

Add to your .mcp.json:

{
  "mcpServers": {
    "nano-banana": {
      "command": "uvx",
      "args": ["gemini-image-mcp"]
    }
  }
}

Or configure in ~/.claude/settings.json:

{
  "env": {
    "GEMINI_API_KEY": "your-api-key-here"
  }
}

Tools

generate_image

Generate images with automatic model selection.

mcp__nano-banana__generate_image

Parameters:

  • prompt (required): Image description
  • model_tier: "flash", "pro", or "auto" (default: auto)
  • aspect_ratio: "1:1", "16:9", "9:16", etc. (default: 1:1)
  • thinking_level: "LOW" or "HIGH" (Pro only)
  • use_grounding: Enable Google Search grounding (Pro only)
  • safety_level: "STRICT", "MODERATE", "PERMISSIVE", "OFF"
  • seed: Integer for reproducible results
  • output_path: File path to save image

list_files

List uploaded files.

mcp__nano-banana__list_files

upload_file

Upload a file for use as reference.

mcp__nano-banana__upload_file

delete_file

Delete an uploaded file.

mcp__nano-banana__delete_file

Model Selection

Pro model auto-selects for prompts containing:

  • professional, 4k, high quality, detailed, photorealistic
  • ultra, premium, studio, commercial, product photo

Flash model auto-selects for prompts containing:

  • quick, fast, sketch, draft, concept, rough
  • preview, test, iterate, simple, basic

Development

# Clone and setup
git clone https://github.com/channel47/mcps.git
cd mcps/nano-banana
uv venv && source .venv/bin/activate
uv pip install -e ".[dev]"

# Run tests
pytest tests/ -v

# Format code
ruff format src/ tests/
ruff check src/ tests/

Links

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

gemini_image_mcp-1.1.1.tar.gz (13.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

gemini_image_mcp-1.1.1-py3-none-any.whl (12.6 kB view details)

Uploaded Python 3

File details

Details for the file gemini_image_mcp-1.1.1.tar.gz.

File metadata

  • Download URL: gemini_image_mcp-1.1.1.tar.gz
  • Upload date:
  • Size: 13.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for gemini_image_mcp-1.1.1.tar.gz
Algorithm Hash digest
SHA256 eb21c1cafafb6539c11e37e9db26e773b43a20c30289eb0fcb55510c1b16b6be
MD5 f73307414e308b7aed94f6edac96d217
BLAKE2b-256 66a33fb5469a16f3303dc45b722fee90565f3c87d8b800780716274a0e29c2d5

See more details on using hashes here.

Provenance

The following attestation bundles were made for gemini_image_mcp-1.1.1.tar.gz:

Publisher: publish.yml on channel47/mcps

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file gemini_image_mcp-1.1.1-py3-none-any.whl.

File metadata

File hashes

Hashes for gemini_image_mcp-1.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 9a7a6032889a21177d9eb0ca94ec66e0a31af8dfde8a3b5ab7c83aa0207f4cc5
MD5 631c84b34e3c8a4a6975316ea82ff104
BLAKE2b-256 08942f2237be21270daff857b28a90fcd688b49e3b79c65da0e9043b52510fac

See more details on using hashes here.

Provenance

The following attestation bundles were made for gemini_image_mcp-1.1.1-py3-none-any.whl:

Publisher: publish.yml on channel47/mcps

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page