ParaView MCP
Integrates multimodal large language models with ParaView to enable the creation and manipulation of scientific visualizations using natural language and visual inputs. It features visual feedback capabilities for iterative refinement, making advance
What is ParaView MCP?
ParaView MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to integrates multimodal large language models with paraview to enable the creation and manipulation of scientific visualizations using natural language and visual inputs. it features visual feedback cap...
Integrates multimodal large language models with ParaView to enable the creation and manipulation of scientific visualizations using natural language and visual inputs. It features visual feedback capabilities for iterative refinement, making advance
This server falls under the Data Science & ML category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Integrates multimodal large language models with ParaView to
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx paraview-mcpConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use ParaView MCP
ParaView-MCP bridges multimodal large language models with ParaView, the open-source scientific visualization platform, enabling users to create, configure, and iteratively refine complex 3D visualizations through natural language and visual feedback. Instead of manually navigating ParaView's GUI, researchers and engineers can describe the visualization they want in plain text and have the AI generate the appropriate ParaView Python pipeline scripts. The visual feedback loop lets the model observe screenshots of the viewport and make iterative adjustments until the result matches the user's intent.
Prerequisites
- Anaconda or Miniconda installed for environment management
- ParaView 5.12+ available via conda-forge (paraview package)
- Python 3.10 in the conda environment
- Claude Desktop configured as the MCP client
- Scientific data files (VTK, VTU, CSV, etc.) accessible on disk
Clone the repository
Clone the ParaView-MCP repository from the LLNL GitHub organization to get the MCP server script and example configurations.
git clone https://github.com/LLNL/paraview_mcp.git
cd paraview_mcpCreate the conda environment
Create a dedicated conda environment with Python 3.10 to avoid conflicts with system Python installations.
conda create -n paraview_mcp python=3.10
conda activate paraview_mcpInstall ParaView and dependencies
Install ParaView from conda-forge along with the MCP SDK and httpx for HTTP communication between the server and client.
conda install conda-forge::paraview
conda install mcp[cli] httpxStart the ParaView server
Launch ParaView's pvserver in multi-client mode so the MCP server can connect to it. Keep this terminal window open while using the integration.
python pvserver --multi-clientsConnect the ParaView GUI
Open the ParaView desktop application, go to File → Connect, and connect to the running pvserver (typically localhost:11111). This establishes the visualization viewport.
Configure Claude Desktop
Add the ParaView MCP server to your Claude Desktop config, pointing to the Python executable in your conda environment and the path to paraview_mcp_server.py.
{
"mcpServers": {
"ParaView": {
"command": "/path/to/conda/envs/paraview_mcp/bin/python",
"args": ["/path/to/paraview_mcp/paraview_mcp_server.py"]
}
}
}Start Claude Desktop and create a visualization
Launch Claude Desktop after saving the config. Ask Claude to create or modify a ParaView visualization — it will generate Python pipeline scripts, observe the viewport via screenshots, and refine the result.
ParaView MCP Examples
Client configuration
Claude Desktop configuration for the ParaView-MCP server using the conda environment's Python executable.
{
"mcpServers": {
"ParaView": {
"command": "/Users/you/miniconda3/envs/paraview_mcp/bin/python",
"args": ["/Users/you/paraview_mcp/paraview_mcp_server.py"]
}
}
}Prompts to try
Example prompts for creating and iterating on scientific visualizations in ParaView via natural language.
- "Load the file simulation.vtu and render it with a blue-to-red color scale based on pressure values."
- "Add a clip plane at z=0 to the current visualization to see the internal structure."
- "Change the opacity of the volume rendering to 0.7 so I can see through the surface."
- "Apply a streamline filter using the velocity field and show me the result."
- "The current color scale is hard to read — switch to a perceptually uniform colormap like viridis."Troubleshooting ParaView MCP
ParaView GUI shows blank viewport or fails to connect to pvserver
Make sure pvserver is running before opening the ParaView GUI and connecting. Use the default port (11111) and connect via File → Connect → localhost:11111. If pvserver crashed, restart it and reconnect.
Synchronization issues or display artifacts in recent ParaView versions
The current implementation relies on pvserver client synchronization which has been deprecated in newer ParaView releases. Use ParaView 5.12 installed from conda-forge for best compatibility, as noted in the project README.
Claude Desktop cannot find the Python executable or MCP server script
Use the full absolute path to the Python binary inside your conda environment (e.g., ~/miniconda3/envs/paraview_mcp/bin/python) and the full path to paraview_mcp_server.py. Relative paths do not work in Claude Desktop's MCP config.
Frequently Asked Questions about ParaView MCP
What is ParaView MCP?
ParaView MCP is a Model Context Protocol (MCP) server that integrates multimodal large language models with paraview to enable the creation and manipulation of scientific visualizations using natural language and visual inputs. it features visual feedback capabilities for iterative refinement, making advance It connects AI assistants to external tools and data sources through a standardized interface.
How do I install ParaView MCP?
Follow the installation instructions on the ParaView MCP GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with ParaView MCP?
ParaView MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is ParaView MCP free to use?
Yes, ParaView MCP is open source and available under the BSD 3-Clause license. You can use it freely in both personal and commercial projects.
ParaView MCP Alternatives — Similar Data Science & ML Servers
Looking for alternatives to ParaView MCP? Here are other popular data science & ml servers you can use with Claude, Cursor, and VS Code.
Ultrarag
★ 5.6kA 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
★ 2.1kAix-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
★ 1.7kOpen 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
★ 1.5kBridges 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.
Set Up ParaView MCP 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 ParaView MCP?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.