Blender AI MCP

v1.0.0APIsstable

Production-shaped MCP server for Blender with goal-first routing, curated tools, deterministic verification, and vision-assisted 3D modeling workflows.

3d-modelingai-agentsai-toolsblender-pythoncomputer-graphics
Share:
26
Stars
0
Downloads
0
Weekly
0/5

What is Blender AI MCP?

Blender AI MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to production-shaped mcp server for blender with goal-first routing, curated tools, deterministic verification, and vision-assisted 3d modeling workflows.

Production-shaped MCP server for Blender with goal-first routing, curated tools, deterministic verification, and vision-assisted 3D modeling workflows.

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

Features

  • Production-shaped MCP server for Blender with goal-first rou

Use Cases

3D modeling automation
Material and blueprint control
VFX workflows
PatrykIti

Maintainer

LicenseApache-2.0
Languagepython
Versionv1.0.0
UpdatedMay 18, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx blender-ai

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 Blender AI MCP

The Blender AI MCP server is a production-grade MCP integration for Blender 5.0+ that enables AI-driven 3D modeling through goal-first routing, curated macro tools, deterministic spatial verification, and vision-assisted workflows. It runs as a Docker container (or locally via Python) and communicates with a Blender addon over a local RPC connection on port 8765, exposing tools for scene analysis, spatial measurement, high-level modeling macros, and iterative reference-based verification. 3D artists, technical directors, and developers use it to delegate repetitive modeling tasks to an AI agent while maintaining precise control over geometry and scene structure.

Prerequisites

  • Blender 5.0 or later installed (4.0+ supported for the addon)
  • Python 3.11 or later if running locally without Docker
  • Docker installed if using the containerized deployment (recommended)
  • An MCP-compatible client such as Claude Desktop
  • The blender-ai-mcp Blender addon installed and enabled in Blender (starts RPC server on port 8765)
1

Install the Blender addon

Download blender_ai_mcp.zip from the GitHub Releases page. In Blender, go to Edit → Preferences → Add-ons, click Install, select the zip file, and enable the addon. It will start an RPC server on port 8765.

2

Pull the Docker image

Pull the pre-built Docker image from GitHub Container Registry.

docker pull ghcr.io/patrykiti/blender-ai-mcp:latest
3

Run the MCP server container

Start the container in stdio mode for Claude Desktop. The BLENDER_RPC_HOST should be host.docker.internal on macOS/Windows. On Linux, use --network host and 127.0.0.1.

docker run -i --rm \
  -v /tmp:/tmp \
  -e BLENDER_AI_TMP_INTERNAL_DIR=/tmp \
  -e BLENDER_AI_TMP_EXTERNAL_DIR=/tmp \
  -e ROUTER_ENABLED=true \
  -e MCP_SURFACE_PROFILE=llm-guided \
  -e BLENDER_RPC_HOST=host.docker.internal \
  ghcr.io/patrykiti/blender-ai-mcp:latest
4

Add to Claude Desktop configuration

Configure Claude Desktop to launch the Docker container as the MCP server.

{
  "mcpServers": {
    "blender-ai": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-v", "/tmp:/tmp",
        "-e", "BLENDER_AI_TMP_INTERNAL_DIR=/tmp",
        "-e", "BLENDER_AI_TMP_EXTERNAL_DIR=/tmp",
        "-e", "ROUTER_ENABLED=true",
        "-e", "MCP_SURFACE_PROFILE=llm-guided",
        "-e", "BLENDER_RPC_HOST=host.docker.internal",
        "ghcr.io/patrykiti/blender-ai-mcp:latest"
      ]
    }
  }
}
5

Set a modeling goal and begin working

In Claude, start by setting a goal with the router_set_goal tool, then use browse_workflows or search_tools to discover relevant macro tools for your task.

Blender AI MCP Examples

Client configuration

Example claude_desktop_config.json using Docker to run the Blender AI MCP server in llm-guided profile mode.

{
  "mcpServers": {
    "blender-ai": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-v", "/tmp:/tmp",
        "-e", "BLENDER_AI_TMP_INTERNAL_DIR=/tmp",
        "-e", "BLENDER_AI_TMP_EXTERNAL_DIR=/tmp",
        "-e", "ROUTER_ENABLED=true",
        "-e", "MCP_SURFACE_PROFILE=llm-guided",
        "-e", "BLENDER_RPC_HOST=host.docker.internal",
        "ghcr.io/patrykiti/blender-ai-mcp:latest"
      ]
    }
  }
}

Prompts to try

Example modeling tasks and instructions to use with Claude and the Blender AI MCP server.

- "Set the goal to model a sci-fi control panel with buttons and screens, then browse available workflows"
- "Analyze the current scene structure and show me the scope graph of objects"
- "Attach the door_handle object to the surface of the door mesh"
- "Apply bevel and subdivision finishing to the selected object"
- "Measure the gap between the two objects and assert they are in contact"
- "Compare the current stage against the reference image and identify differences"

Troubleshooting Blender AI MCP

Cannot connect to Blender RPC server on port 8765

Ensure the blender-ai-mcp addon is installed and enabled in Blender — it must be running before the MCP container starts. On Linux, replace host.docker.internal with 127.0.0.1 and add --network host to the Docker run command. Verify no firewall is blocking port 8765.

Docker container exits immediately in stdio mode

Make sure you include the -i flag (interactive/stdin) in the docker run command. Without -i, the container exits when there is no stdin. The MCP protocol uses stdio for communication and requires an open stdin pipe.

Macro tools fail with spatial assertion errors

Run scene_scope_graph and scene_view_diagnostics before invoking macros to ensure the target objects exist and are visible. Macro tools like macro_attach_part_to_surface require specific object names that match exactly what is in the Blender scene.

Frequently Asked Questions about Blender AI MCP

What is Blender AI MCP?

Blender AI MCP is a Model Context Protocol (MCP) server that production-shaped mcp server for blender with goal-first routing, curated tools, deterministic verification, and vision-assisted 3d modeling workflows. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Blender AI MCP?

Follow the installation instructions on the Blender AI MCP GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Blender AI MCP?

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

Is Blender AI MCP free to use?

Yes, Blender AI MCP is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Browse More APIs MCP Servers

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

Quick Config Preview

{ "mcpServers": { "blender-ai": { "command": "npx", "args": ["-y", "blender-ai"] } } }

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

Read the full setup guide →

Ready to use Blender AI MCP?

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