Use This MCP server To

Control Blender using natural language prompts Automate 3D modeling tasks with AI commands Integrate local AI models into Blender workflows Download and use PolyHaven assets via AI prompts Query scene and object information programmatically Extend Blender functionality with AI-driven tools

README

blender-open-mcp

blender-open-mcp is an open source project that integrates Blender with local AI models (via Ollama) using the Model Context Protocol (MCP). This allows you to control Blender using natural language prompts, leveraging the power of AI to assist with 3D modeling tasks.

Features

Control Blender with Natural Language: Send prompts to a locally running Ollama model to perform actions in Blender.
MCP Integration: Uses the Model Context Protocol for structured communication between the AI model and Blender.
Ollama Support: Designed to work with Ollama for easy local model management.
Blender Add-on: Includes a Blender add-on to provide a user interface and handle communication with the server.
PolyHaven Integration (Optional): Download and use assets (HDRIs, textures, models) from PolyHaven directly within Blender via AI prompts.
Basic 3D Operations:
- Get Scene and Object Info
- Create Primitives
- Modify and delete objects
- Apply materials
Render Support: Render images using the tool and retrieve information based on the output.

Installation

Prerequisites

Blender: Blender 3.0 or later. Download from blender.org.
Ollama: Install from ollama.com, following OS-specific instructions.
Python: Python 3.10 or later.
uv: Install using pip install uv.
Git: Required for cloning the repository.

Installation Steps

Clone the Repository:

git clone https://github.com/dhakalnirajan/blender-open-mcp.git
cd blender-open-mcp

Create and Activate a Virtual Environment (Recommended):

uv venv
source .venv/bin/activate  # On Linux/macOS
.venv\Scripts\activate  # On Windows

Install Dependencies:
```
uv pip install -e .
```
Install the Blender Add-on:
- Open Blender.
- Go to Edit -> Preferences -> Add-ons.
- Click Install....
- Select the addon.py file from the blender-open-mcp directory.
- Enable the "Blender MCP" add-on.
Download an Ollama Model (if not already installed):
```
ollama run llama3.2
```
(Other models like Gemma3 can also be used.)

Setup

Start the Ollama Server: Ensure Ollama is running in the background.

Start the MCP Server:

blender-mcp

Or,

python src/blender_open_mcp/server.py

By default, it listens on http://0.0.0.0:8000, but you can modify settings:

blender-mcp --host 127.0.0.1 --port 8001 --ollama-url http://localhost:11434 --ollama-model llama3.2

Start the Blender Add-on Server:
- Open Blender and the 3D Viewport.
- Press N to open the sidebar.
- Find the "Blender MCP" panel.
- Click "Start MCP Server".

Usage

Interact with blender-open-mcp using the mcp command-line tool:

Example Commands

Basic Prompt:

mcp prompt "Hello BlenderMCP!" --host http://localhost:8000

Get Scene Information:

mcp tool get_scene_info --host http://localhost:8000

Create a Cube:

mcp prompt "Create a cube named 'my_cube'." --host http://localhost:8000

Render an Image:

mcp prompt "Render the image." --host http://localhost:8000

Using PolyHaven (if enabled):

mcp prompt "Download a texture from PolyHaven." --host http://localhost:8000

Available Tools

Tool Name	Description	Parameters
`get_scene_info`	Retrieves scene details.	None
`get_object_info`	Retrieves information about an object.	`object_name` (str)
`create_object`	Creates a 3D object.	`type`, `name`, `location`, `rotation`, `scale`
`modify_object`	Modifies an object’s properties.	`name`, `location`, `rotation`, `scale`, `visible`
`delete_object`	Deletes an object.	`name` (str)
`set_material`	Assigns a material to an object.	`object_name`, `material_name`, `color`
`render_image`	Renders an image.	`file_path` (str)
`execute_blender_code`	Executes Python code in Blender.	`code` (str)
`get_polyhaven_categories`	Lists PolyHaven asset categories.	`asset_type` (str)
`search_polyhaven_assets`	Searches PolyHaven assets.	`asset_type`, `categories`
`download_polyhaven_asset`	Downloads a PolyHaven asset.	`asset_id`, `asset_type`, `resolution`, `file_format`
`set_texture`	Applies a downloaded texture.	`object_name`, `texture_id`
`set_ollama_model`	Sets the Ollama model.	`model_name` (str)
`set_ollama_url`	Sets the Ollama server URL.	`url` (str)
`get_ollama_models`	Lists available Ollama models.	None

Troubleshooting

If you encounter issues:

Ensure Ollama and the blender-open-mcp server are running.
Check Blender’s add-on settings.
Verify command-line arguments.
Refer to logs for error details.

For further assistance, visit the GitHub Issues page.

Happy Blending with AI! 🚀

blender-open-mcp FAQ

How do I install blender-open-mcp?

Install the Blender add-on included with blender-open-mcp and run the MCP server locally with Ollama models configured.

Can I use blender-open-mcp without Ollama?

No, blender-open-mcp is designed specifically to work with Ollama for local AI model management.

Does blender-open-mcp support remote AI models?

It primarily supports local models via Ollama, but MCP architecture could allow extensions for remote models.

How does blender-open-mcp handle asset downloads from PolyHaven?

It integrates PolyHaven API calls to download HDRIs, textures, and models directly into Blender via AI prompts.

Is the communication between Blender and AI models secure?

Yes, MCP ensures scoped, secure, and observable interactions between Blender and AI models.

What Blender versions are supported?

It supports Blender versions compatible with the add-on, typically recent stable releases.

Can I customize the AI prompts or commands?

Yes, users can craft custom natural language prompts to control Blender operations through the MCP server.

How does MCP improve Blender automation?

MCP standardizes AI communication, enabling multi-step reasoning and complex task execution within Blender.