z-image-studio

Name: z-image-studio
Rating: 2.5 (115 reviews)
Author: iconben

by iconben·★ 115·Score 49

A comprehensive MCP server for Z-Image-Turbo text-to-image generation with CLI, Web UI, and multiple GPU optimizations.

ai-llmdeveloper-toolsmedia

Forks

Open issues

this month

Last commit

2d ago

Indexed

Overview

Z-Image Studio provides a complete solution for text-to-image generation using the Z-Image-Turbo model. It offers multiple interfaces: a command-line interface for direct image generation, a modern web UI with multilingual support, and an MCP server that allows AI agents to access image generation capabilities. The application is optimized for various hardware configurations including NVIDIA GPUs, AMD GPUs on Linux, Intel XPU, and Apple Silicon with automatic fallbacks to CPU. The MCP server supports multiple transport protocols including stdio, SSE, and Streamable HTTP, ensuring compatibility with various AI client implementations.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:AI agents needing to generate images from text descriptions

you:Automated content creation workflows requiring visual assets

you:Local development environments with image generation capabilities

you:What GPU configurations are supported?

you:How can I use the MCP server with different AI agents?

When to choose this

Choose this MCP server when you need local image generation with support for multiple hardware platforms and want both CLI and web interfaces alongside MCP integration.

When NOT to choose this

Don't choose this if you need cloud-based image generation without local hardware requirements or if you need integration with image generation models other than Z-Image-Turbo.

Tools this server exposes

3 tools extracted from the README

generate
Generate an image from a text prompt
list_models
List available image generation models
list_history
View past image generation history

Comparable tools

stable-diffusion-webuicomfyuikohya_ss

Installation

**Using uv (recommended):**

uv tool install git+https://github.com/iconben/z-image-studio.git

**Using pip:**

pip install z-image-studio

**Global CLI Installation:**

uv tool install z-image-studio
# Update:
uv tool upgrade z-image-studio

**MCP Server Configuration:** After installation, the MCP server can be accessed via:

stdio: zimg mcp or zimg-mcp
SSE: /mcp-sse
Streamable HTTP: /mcp

For Claude Desktop, add to config.json:

{
  "mcpServers": {
    "z-image-studio": {
      "command": "zimg",
      "args": ["mcp"]
    }
  }
}

FAQ

What GPU configurations are supported?: The application supports NVIDIA (CUDA), AMD on Linux (ROCm), Intel XPU, and Apple Silicon (MPS), with automatic fallback to CPU if no compatible GPU is detected.
How can I use the MCP server with different AI agents?: The MCP server supports multiple transport protocols: stdio via `zimg mcp`, SSE at `/mcp-sse`, and Streamable HTTP at `/mcp`. Try the Streamable HTTP first for optimal performance.

Compare z-image-studio with

z-image-studio vs ultimate_mcp_server z-image-studio vs mcp-server-chart z-image-studio vs everything z-image-studio vs filesystem z-image-studio vs time

GitHub →

Last updated 2026-05-17 · Auto-generated from public README + GitHub signals.