blender_fleet
Blender MCP Fleet 原型
本原型对外暴露一个 MCP 调度器,并将各 ai 作业独立扇出到隔离、无头的 Blender 进程。它面向多个智能体:每个智能体提交一个作业,每个作业都有自己的目录、.blend、预览、日志和 Blender 进程。
为什么这样设计
当前安装的 ahujasid/n 插件是一个 GUI TCP 桥接器,默认只使用一个固定端口。工作池避免了端口冲突和不同项目之间的交叉污染。它也使故障可以恢复:一个工作进程失败,只是一个作业失败。
Hmm wait — I just reproduced "ahujasid/n"? No. The source is ahujasid/blender-mcp. Let me be careful — my written final should use the exact content. Reread source: "The currently installed ahujasid/blender-mcp add-on..." — must keep. Good.
I should be careful not to mistype. Let me produce final now# Blender MCP Fleet 原型
本原型对外暴露一个 MCP 调度器,并将独立的模型作业分发到隔离、无头的 Blender 进程。它面向多个智能体:每个智能体提交一个作业,每个作业都有自己的目录、.blend、预览、日志和 Blender 进程。
Related MCP server: Blender MCP Server
为什么这样设计
当前安装的 ahujasid/blender-mcp 插件是一个 GUI TCP 桥接器,默认使用一个固定端口。工作池避免了端口冲突,也避免了不同项目之间的交叉污染。它还让故障可以恢复:一个失败的工作进程只是一个失败的作业。
运行冒烟测试
从 PowerShell 运行:
Set-Location D:\blender-mcp-fleet\prototype\fleet
python .\smoke_test.py该测试使用两个工作进程提交四个模型,因此会同时运行两个 Blender 作业。结果会写入 D:\blender-mcp-fleet\prototype\jobs。
运行 MCP 调度器
MCP 依赖可以通过 uv 安装到一个本地环境中:
Set-Location D:\blender-mcp-fleet\prototype
uv venv --python 3.11 .venv
uv pip install --python .\.venv\Scripts\python.exe "mcp[cli]==1.29.0"然后,将它注册为用于测试的单独 MCP 条目;它不会替换当前的 blender 条目:
[mcp_servers.blender_fleet]
command = "D:\\blender-mcp-fleet\\prototype\\.venv\\Scripts\\python.exe"
args = ["D:\\blender-mcp-fleet\\prototype\\fleet\\mcp_server.py"]
[mcp_servers.blender_fleet.env]
BLENDER_FLEET_PARALLEL = "2"暴露的工具包括 submit_model_job、get_job_status、wait_for_job、list_model_jobs 和 fleet_info。
安全的预设类型包括 cube、sphere、monkey、tower、robot、spaceship、castle 和 solar_system。最后三个类型适合用来做更真实的多智能体冒烟测试,因为它们会创建多个网格、材质、灯光和修改器。
想在不修改 Codex 配置的情况下测试实际的 MCP stdio 握手:
D:\blender-mcp-fleet\prototype\.venv\Scripts\python.exe .\fleet\mcp_smoke.py安全边界
第一版只接受一个小型的已列入白名单的模型类型集。它不会执行任意由智能体提供的 Python 代码。后续版本可以加入一个经过评审的 recipe/script 模式,支持按作业设置超时时间和限定工作区的白名单。
This server cannot be deployed
Maintenance
Related MCP Connectors
Blender cloud GPU render farm for AI agents (MCP/API): .blend, bpy, glTF/FBX/USD in; frames/MP4 out
Blender-as-a-service for agents: search 3D assets, run Blender Python, or brief the studio agent.
Durable background job execution, async task scheduling, and state persistence for AI agents.
Cloud Blender for AI agents: build, inspect, render and animate 3D scenes over remote MCP. Keep editable .blend projects and export GLB or STL. Make your first 3D asset free—30 compute minutes per UTC month, no credit card. First 100 eligible Free account owners to use all 30 minutes in one UTC month by December 31, 2026 UTC receive one calendar month of beta free: 4 compute hours per UTC month, 2 concurrent workers per deployment, 2 deployments, 2048×2048 renders at up to 256 samples, and 10 GiB storage. Existing usage counts toward the 4-hour allowance. Limited to 100 rewards; expires January 1, 2027 00:00 UTC. Free resumes afterward without an automatic charge. Examples, eligibility and terms: https://sceneplane.online/launch?utm_source=glama&utm_medium=directory&utm_campaign=first100
Related MCP Servers
- AlicenseNot gradedqualityNot gradedmaintenanceEnables AI agents to control Blender 3D through 27 tools for scene manipulation, materials, modifiers, animation, and rendering, with native Python integration built in Rust.MIT
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to control Blender 3D software through natural language commands, supporting object creation, manipulation, materials, rendering, and scene management with 22 tools organized across 6 categories.-
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to fully control Blender through 50+ tools for 3D modeling, animation, materials, and scene management via HTTP endpoints.-
- AlicenseCqualityAmaintenanceEnables AI agents to control Blender through 150+ tools across 24 modules, including modeling, materials, lighting, animation, rendering, mesh quality analysis, and visual feedback, plus expert prompts and goal-first workflow management.209MIT