Z-Image Studio
A local text-to-image generation server that enables AI agents to generate images, manage models, and browse history using the Z-Image-Turbo model. It supports hardware acceleration across NVIDIA, Apple Silicon, and AMD GPUs via multiple MCP transpor
What is Z-Image Studio?
Z-Image Studio is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to local text-to-image generation server that enables ai agents to generate images, manage models, and browse history using the z-image-turbo model. it supports hardware acceleration across nvidia, apple...
A local text-to-image generation server that enables AI agents to generate images, manage models, and browse history using the Z-Image-Turbo model. It supports hardware acceleration across NVIDIA, Apple Silicon, and AMD GPUs via multiple MCP transpor
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A local text-to-image generation server that enables AI agen
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx z-image-studioConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Z-Image Studio
Z-Image Studio is a local text-to-image generation server powered by the Tongyi-MAI/Z-Image-Turbo diffusion model from Hugging Face. It runs entirely on your machine and exposes three MCP tools — generate, list_models, and list_history — so AI agents can generate images, manage models, and browse generation history without any cloud API calls. It supports hardware acceleration on NVIDIA CUDA, Apple Silicon (MPS), AMD ROCm, and Intel XPU, making it a fully offline, privacy-preserving alternative to cloud image generation APIs.
Prerequisites
- Python 3.11+ (Python 3.12+ supported, but torch.compile is disabled by default on 3.12)
- pip or uv package manager
- A supported GPU: NVIDIA (CUDA), Apple Silicon (MPS), AMD (ROCm on Linux), or Intel XPU — or CPU as fallback
- Sufficient disk space for model weights (~few GB) which are downloaded from Hugging Face on first run
- An MCP-compatible client such as Claude Desktop or Claude Code
Install Z-Image Studio
Install via pip or uv. The uv tool install method is recommended as it creates an isolated environment automatically.
# Recommended: via uv tool (isolated install)
uv tool install git+https://github.com/iconben/z-image-studio.git
# Or via pip
pip install z-image-studioStart the MCP server
Run `zimg-mcp` to launch the MCP server in stdio mode. On first run, the Z-Image-Turbo model weights are downloaded from Hugging Face automatically.
zimg-mcp
# First run downloads model weights — this may take several minutes
# depending on your internet connection speedOptional: test image generation via CLI
Verify the installation works by generating a test image from the command line before configuring the MCP client.
# Basic generation test
zimg generate "A futuristic city with neon lights"
# With options
zimg gen "A majestic dragon" --seed 12345 --steps 20 --precision q8
# List available models
zimg modelsConfigure your MCP client
Add Z-Image Studio to your MCP client configuration. The server exposes three tools: generate, list_models, and list_history.
# Tools exposed:
# generate - Generate an image from a text prompt
# list_models - List available model variants
# list_history - Browse previously generated imagesOptional: launch the web UI
Z-Image Studio also includes a web interface for browsing history and generating images visually.
# Start the web UI on port 8000
zimg serve --host 0.0.0.0 --port 8000
# Open http://localhost:8000 in your browserZ-Image Studio Examples
Client configuration (Claude Desktop)
Configure Claude Desktop to launch the Z-Image Studio MCP server. The zimg-mcp command must be on your PATH after installation.
{
"mcpServers": {
"z-image-studio": {
"command": "zimg-mcp",
"env": {
"Z_IMAGE_STUDIO_OUTPUT_DIR": "/path/to/output/images"
}
}
}
}Prompts to try
Use these prompts to generate images, check available models, and browse history through your AI assistant.
- "Generate an image of a serene mountain lake at sunrise."
- "Create a photorealistic portrait of an astronaut on Mars."
- "Generate a pixel art style cityscape at night."
- "What image generation models are available?"
- "Show me the last 10 images I generated."
- "Generate an image with seed 42 so I can reproduce it later."Troubleshooting Z-Image Studio
Model download fails or takes too long on first run
The Z-Image-Turbo model weights are downloaded from Hugging Face on first run. Ensure you have a stable internet connection and sufficient disk space. If the download fails, retry running `zimg-mcp` — it resumes partial downloads.
Generated images are black or all the same colour
This is typically caused by the safety filter intercepting outputs. Z-Image Studio disables the safety filter by default to prevent false positives, but verify this in your config. Also try a different precision setting: `--precision q8` or `--precision q4` for lower memory usage.
Out of memory errors on GPU during generation
Reduce image dimensions (default is 1280×720 — try 512×512 for testing), use a lower precision model (`--precision q4`), or set `ZIMAGE_ENABLE_TORCH_COMPILE=0` to disable torch compilation. Automatic attention slicing is enabled by default to reduce VRAM usage.
Frequently Asked Questions about Z-Image Studio
What is Z-Image Studio?
Z-Image Studio is a Model Context Protocol (MCP) server that local text-to-image generation server that enables ai agents to generate images, manage models, and browse history using the z-image-turbo model. it supports hardware acceleration across nvidia, apple silicon, and amd gpus via multiple mcp transpor It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Z-Image Studio?
Follow the installation instructions on the Z-Image Studio GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Z-Image Studio?
Z-Image Studio works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Z-Image Studio free to use?
Yes, Z-Image Studio is open source and available under the Apache 2.0 license. You can use it freely in both personal and commercial projects.
Z-Image Studio Alternatives — Similar Developer Tools Servers
Looking for alternatives to Z-Image Studio? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Z-Image Studio in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Z-Image Studio?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.