https://github.com/FishWoWater/trellis_mcp
Model Context Protocol(MCP) for TRELLIS(SOTA text-to-3d/image-to-3d) models
https://github.com/FishWoWater/trellis_mcp
blender image-to-3d mcp mcp-server text-to-3d trellis
Last synced: 9 months ago
JSON representation
Model Context Protocol(MCP) for TRELLIS(SOTA text-to-3d/image-to-3d) models
- Host: GitHub
- URL: https://github.com/FishWoWater/trellis_mcp
- Owner: FishWoWater
- Created: 2025-03-25T15:51:52.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-04-06T10:13:32.000Z (about 1 year ago)
- Last Synced: 2025-04-06T11:22:55.925Z (about 1 year ago)
- Topics: blender, image-to-3d, mcp, mcp-server, text-to-3d, trellis
- Language: Python
- Homepage:
- Size: 2.61 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-mcp-servers - **trellis_mcp** - Model Context Protocol(MCP) for TRELLIS(SOTA text-to-3d/image-to-3d) models `python` `blender` `image-to-3d` `mcp` `mcp-server` `pip install git+https://github.com/FishWoWater/trellis_mcp` (🤖 AI/ML)
README
# Trellis MCP Server
Trellis MCP provides an interface between AI assistants and [Trellis](https://github.com/microsoft/TRELLIS) via [Model Context Protocol (MCP)](https://modelcontextprotocol.io).
### Disclaimer
This project shows a very **minimal** integration of MCP with [Trellis](https://github.com/microsoft/TRELLIS): a lightweight and opensource text-to-3d/image-to-3d 3DAIGC model. Compared with existing [rodin integration in blender-mcp](https://github.com/ahujasid/blender-mcp) and [tripo integration](https://github.com/VAST-AI-Research/tripo-mcp), it has following advantages:
* **Faster and memory-efficient**: You can deploy TRELLIS **locally with only 8GPU+ VRAM**, while can generate a textured mesh from text in only *~15s*(10s with more vram).
* **FREE**: You DON'T have to pay expensive API from Rodin/Meshy/Tripo.
**BUT IT HAS FOLLOWING LIMITATIONS:**
* Trellis is open-source and there is no off-the-shelf API model providers, you have to deploy it by yourself (refer to [README](https://github.com/FishWoWater/TRELLIS/blob/dev/README_api.md)).
* The API/Prompt has NOT been fully tested/tuned, may suffer from stability issues.
So use it at your own risk.
## Demo
> A minimal demo for generating a single object, more complicated prompt with blender-mcp is under tuning.
## Features
- [x] Generate 3D asset from natural language(**TEXT**) using Trellis API and import into blender
- [ ] Generate texture/materials from natural language(**TEXT**) for a given 3D mesh using Trellis API and import into blender
## Roadmap
### Prerequisites
- Python 3.10+
- [Blender](https://www.blender.org/download/)
- [Trellis Blender Addon](https://github.com/FishWoWater/trellis_blender)
- [Trellis API Backend](https://github.com/FishWoWater/TRELLIS)
- Claude / Cursor(SUGGESTED) / Windsurf
### Installation
#### 1. Trellis blender addon
1. Download Trellis Blender Addon from [here](https://github.com/FishWoWater/trellis_blender)
2. Open Blender -> Edit -> Preferences -> Add-ons -> Install from file -> Select the downloaded addon -> Install
3. In 3D Viewport -> View3D > Sidebar > TRELLIS -> Start MCP Server
#### 2. Configure API backend
> As trellis is a free open-source text-to-3d model, you need to deploy your own trellis API backend with reference to [README](https://github.com/FishWoWater/TRELLIS/blob/dev/README_api.md)
``` shell
# clone an API fork of trellis
git clone https://github.com/FishWoWater/TRELLIS
cd TRELLIS
# EDIT BACKEND URL in trellis_api/config.py
# configure the # of text workers and start ai worker
# no need for image workers
python trellis_api/ai_worker.py --text-workers-per-gpu 1 --image-workers-per-gpu 0
# start web server
python trellis_api/web_server.py
# or on windows local dev
python trellis_api/web_server_single.py
```
#### 3. Configure the MCP server on Windsurf/Cursor/Claude
```text
{
"mcpServers": {
"trellis-blender": {
"command": "uvx",
"args": [
"trellis-mcp"
]
}
}
}
```
## Acknowledgements
- Backbone and brain: [Trellis](https://github.com/microsoft/TRELLIS)
- Inspiration: [blender-mcp](https://github.com/ahujasid/blender-mcp)
- Borrow a lot of code [Tripo MCP Service](https://github.com/VAST-AI-Research/tripo-mcp)