MCPgee
MCPgee
MCPgeeMCPgee

ComfyUI

v0.2.1•Data Science & ML•stable

MCP server for ComfyUI — text-to-image, variations, img2img refine, upscale, image proxy, and workflow runner.

comfyuicomfyui-apiimage-mcplocal-datalocal-mcp
Share:
320
Stars
0
Downloads
0
Weekly
0/5
View on GitHub

What is ComfyUI?

ComfyUI is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp server for comfyui — text-to-image, variations, img2img refine, upscale, image proxy, and workflow runner.

MCP server for ComfyUI — text-to-image, variations, img2img refine, upscale, image proxy, and workflow runner.

This server falls under the Data Science & ML category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • MCP server for ComfyUI — text-to-image, variations, img2img

Use Cases

Generate images from text prompts.
Perform image upscaling and refinement.
Run custom ComfyUI workflows.
miller-joe

Maintainer

LicenseMIT License
Languagepython
Versionv0.2.1
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux
View SourceBrowse All Servers

Installation

NPM

npx -y @miller-joe/comfyui-mcp

Manual Installation

npx -y @miller-joe/comfyui-mcp

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use ComfyUI

ComfyUI MCP is an MCP server that connects AI assistants to a local or remote ComfyUI instance, exposing 16 tools for text-to-image generation, image variations, img2img refinement, upscaling, ControlNet conditioning, IP-Adapter style transfer, and full custom workflow execution. Rather than using ComfyUI's visual node editor manually, users can ask Claude or another MCP client to generate images with natural language and have the server translate those requests into ComfyUI API calls. It is particularly useful for integrating generative image capabilities into automated pipelines or conversational interfaces.

Prerequisites

  • A running ComfyUI instance accessible at an HTTP URL (default: http://127.0.0.1:8188)
  • At least one Stable Diffusion checkpoint model installed in ComfyUI's models/checkpoints/ directory
  • Node.js 18+ and npx (for the npx transport), or Docker (for the container transport)
  • An MCP-compatible client such as Claude Desktop or Claude Code
1

Start your ComfyUI instance

Launch ComfyUI and confirm it is accessible at its API endpoint. By default this is http://127.0.0.1:8188.

# In your ComfyUI directory:
python main.py --listen
2

Run the comfyui-mcp server with npx

Start the MCP server, pointing it at your ComfyUI instance. Adjust the URL if ComfyUI is on a different host or port.

npx -y @miller-joe/comfyui-mcp --comfyui-url http://127.0.0.1:8188
3

Configure the MCP client

Add the server to your Claude Desktop config file. Pass environment variables to set the ComfyUI URL and the default checkpoint model.

4

Verify available models

Once connected, ask the AI assistant to list available models. This confirms the server can reach ComfyUI and that checkpoints are loaded correctly.

5

Generate a test image

Try a simple text-to-image prompt to confirm the full pipeline works end-to-end before building more complex workflows.

ComfyUI Examples

Client configuration

Register comfyui-mcp in Claude Desktop's config. Set COMFYUI_URL to point to your ComfyUI server and COMFYUI_DEFAULT_CKPT to a model file that exists in your checkpoints folder.

{
  "mcpServers": {
    "comfyui": {
      "command": "npx",
      "args": ["-y", "@miller-joe/comfyui-mcp"],
      "env": {
        "COMFYUI_URL": "http://127.0.0.1:8188",
        "COMFYUI_PUBLIC_URL": "http://127.0.0.1:8188",
        "COMFYUI_DEFAULT_CKPT": "sd_xl_base_1.0.safetensors",
        "MCP_TRANSPORT": "stdio"
      }
    }
  }
}

Prompts to try

Natural language prompts that map to the server's 16 tools.

- "Generate a photorealistic image of a mountain lake at sunset, 1024x768"
- "Create 4 variations of this landscape image with slightly different lighting"
- "Upscale this image to 4x resolution using ComfyUI"
- "List all available checkpoint models loaded in ComfyUI"
- "Run my saved 'portrait-workflow' template with this prompt: elegant woman, studio lighting"

Troubleshooting ComfyUI

Connection refused when the server tries to reach ComfyUI

Make sure ComfyUI is started with '--listen' so it binds to 0.0.0.0 rather than just localhost. If ComfyUI is on a remote machine, use its actual IP address in COMFYUI_URL and ensure port 8188 is open in the firewall.

Image generation fails with 'model not found' or checkpoint errors

Run 'list_models' from the MCP session to see what checkpoints ComfyUI can see. Set COMFYUI_DEFAULT_CKPT to the exact filename returned by that list (including the .safetensors extension). Models must be placed in ComfyUI's models/checkpoints/ directory.

Frequently Asked Questions about ComfyUI

What is ComfyUI?

ComfyUI is a Model Context Protocol (MCP) server that mcp server for comfyui — text-to-image, variations, img2img refine, upscale, image proxy, and workflow runner. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install ComfyUI?

Install via npm with the command: npx -y @miller-joe/comfyui-mcp. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with ComfyUI?

ComfyUI works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is ComfyUI free to use?

Yes, ComfyUI is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Learn More About MCP Servers

Getting Started with MCP

Set up your first MCP server in minutes

MCP Setup Guide

Configure MCP in Claude, Cursor & VS Code

All MCP Tutorials

18+ hands-on guides for developers

MCP FAQ

40+ answers about Model Context Protocol

ComfyUI Alternatives — Similar Data Science & ML Servers

Looking for alternatives to ComfyUI? Here are other popular data science & ml servers you can use with Claude, Cursor, and VS Code.

Ultrarag

★ 5.6k

A Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines

RocketRide

★ 3.1k

📇 🏠 - MCP server that exposes RocketRide AI pipelines as t

Aix Db

★ 2.1k

Aix-DB 基于 LangChain/LangGraph 框架，结合 MCP Skills 多智能体协作架构，实现自然语言到数据洞察的端到端转换。

NeMo Data Designer

★ 1.9k

🎨 NeMo Data Designer: Generate high-quality synthetic data from scratch or from seed data.

PaperBanana

★ 1.7k

Open source implementation and extension of Google Research’s PaperBanana for automated academic figures, diagrams, and research visuals, expanded to new domains like slide generation.

MiniMax

★ 1.5k

Bridges MiniMax AI capabilities to the Model Context Protocol, enabling AI agents to perform image understanding, text-to-image generation, and speech synthesis. It provides a standardized interface for accessing MiniMax's core tools via JSON-RPC.

Browse More Data Science & ML MCP Servers

Explore all data science & ml servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Data Science & MLBrowse All Servers

Quick Config Preview

{ "mcpServers": { "comfyui": { "command": "npx", "args": ["-y", "@miller-joe/comfyui-mcp"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use ComfyUI?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides
Explore All ServersRead Our Guides