Blender add-on connecting 3D scenes to LLMs via an MCP server for AI-driven modeling, lighting, and automation.
Connect ChatGPT, Claude, Gemini, or Ollama to Blender via MCP to build, edit, and inspect 3D scenes.
mcp-for-blender-pro is an open-source Blender extension and MCP server that brings conversational AI into your 3D pipeline. You can inspect scene elements, generate primitive layouts, adjust materials, and delegate tasks to multi-agent chains without leaving your Blender workspace. It is built for 3D generalists, motion designers, and technical artists who want to convert natural language instructions into Blender scene operations.
The tool embeds a Model Context Protocol (MCP) server inside Blender and provides a companion visual Studio window for chat, viewport snapshots, and task tracking. It includes 21 native scene tools for object manipulation, camera placement, Python scripting, node inspection, and asset imports from external libraries like Poly Haven. You can run single AI assistants or configure sequential multi-agent chains containing manager, worker, and reviewer roles.
This project suits 3D artists, technical directors, and developers who want to control Blender via local or cloud LLMs. It is not intended for users looking for zero-input 3D image-to-mesh generators or those operating outside a Windows environment. You also need to supply your own API keys or host a local Ollama server.
Inspect chat history, review task boards, view viewport snapshots, and manage tool approvals in a separate window.
Assign sequential manager, worker, and reviewer roles to break down complex 3D assignments and verify results.
Execute object creation, material assignment, scene queries, Python script runs, and rendering exports automatically.
Search and import assets from Poly Haven, Sketchfab, and Poly Pizza directly through natural language prompts.
Expose Blender tools to external MCP clients such as OpenCode Portable using local Streamable HTTP endpoints.
Free under the MIT license. The software itself carries no subscription cost, but you are responsible for paying API fees charged by providers like OpenAI and Anthropic or supplying hardware for local Ollama instances.
883 open remote AI-training roles, updated November 16, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
You can connect OpenAI, Anthropic Claude, Google Gemini, OpenRouter, OpenCode Zen, local Ollama models, or custom endpoints compatible with OpenAI or Anthropic protocols.
The provided release package targets Windows x64 running Blender 4.2+. Other platforms may require manual building or custom environment setup.
The add-on inspects and validates script calls to block risky operations, but it does not run in a isolated sandbox. You should review operations before confirming approvals.
See Zekai first in Google