invokeai-mcp-server

AI-powered image generation, transformation, and upscaling for Claude Code using your local InvokeAI instance.

## Overview

The InvokeAI MCP Server bridges Claude Code with InvokeAI, enabling seamless AI-assisted image creation directly
from your development environment. Perfect for generating logos, icons, illustrations, and visual assets without
leaving your workflow.

## Features

### Core Capabilities
- **Text-to-Image Generation** - Create images from natural language descriptions using Stable Diffusion 1.5,
SDXL, and custom models
- **Image-to-Image Transformation** - Refine, modify, or stylize existing images with AI guidance
- **LoRA Model Support** - Apply fine-tuned LoRA models for specialized styles (logos, illustrations, vector art)
- **AI-Powered Upscaling** - Enhance images to 2x-4x resolution using state-of-the-art Spandrel models
- **VAE Override** - Fix incompatible built-in VAEs or optimize with specialized VAE models

### Advanced Features
- **Full SDXL Support** - Automatic detection and configuration for SDXL models with dual CLIP encoders
- **Comprehensive Parameter Control** - Fine-tune with width, height, steps, CFG scale, 14+ schedulers, and seeds
- **Queue Management** - Monitor and track InvokeAI processing status
- **Flexible Model Support** - Compatible with SD 1.5, SDXL, and thousands of community models from Civitai and
HuggingFace

## Prerequisites

- **InvokeAI v4.0+** running locally (default: `http://127.0.0.1:9090`)
- **Hardware**: GPU with 4-12GB VRAM depending on model choice
- **Python 3.8+** for non-Docker installation

## Installation

### Via Smithery (Recommended)
```bash
npx @smithery/cli install invokeai --client claude

Via PyPI

pip install invokeai-mcp-server
claude mcp add --scope user invokeai python -m invokeai_mcp_server

Usage Examples

Generate a Logo

Generate a minimalist tech startup logo with blue gradient, geometric shapes, modern flat design

Apply LoRA Styling

Create a professional logo using the logomkrdsxl LoRA model

Refine an Image

Transform logo.png with strength 0.6 to make it more polished and professional

Upscale for Production

Upscale generated_logo.png to high resolution for print

Configuration

When installed via Smithery, you can optionally configure the InvokeAI URL:

npx @smithery/cli install invokeai --client claude \
--config '{"invokeaiUrl":"http://192.168.1.100:9090"}'

Default: http://127.0.0.1:9090

Why InvokeAI MCP Server?

- Workflow Integration - Generate assets without context switching
- Production Quality - Professional-grade Stable Diffusion and SDXL models
- Local & Private - Runs on your hardware, no cloud dependencies
- Highly Customizable - LoRAs, VAE overrides, extensive parameters
- Developer Friendly - Built for programmers creating visual assets

Resources

- PyPI Package: https://pypi.org/project/invokeai-mcp-server/
- GitHub Repository: https://github.com/coinstax/invokeai-mcp-server
- InvokeAI: https://invoke.com/invokeai

License

MIT License - © 2025 CoinsTax LLC