ComfyUI-3D-Pack

Make 3D assets generation in ComfyUI good and convenient as it generates image/video! <br> This is an extensive node suite that enables ComfyUI to process 3D inputs (Mesh & UV Texture, etc.) using cutting edge algorithms (3DGS, NeRF, etc.) and models (InstantMesh, CRM, TripoSR, etc.)

<span style="font-size:1.5em;"> <a href=#Features>Features</a> &mdash; <a href=#install>Install</a> &mdash; <a href=#roadmap>Roadmap</a> &mdash; <a href=#development>Development</a> &mdash; <a href=#tips>Tips</a> &mdash; <a href=#supporters>Supporters</a> </span>

Install:

Can be installed directly from ComfyUI-Manager🚀

Alternatively you can download Comfy3D-WinPortable made by YanWenKun

• Pre-builds are available for:
  • Windows 10/11
  • Python 3.12
  • CUDA 12.4
  • torch 2.5.1+cu124
• install.py will download & install Pre-builds automatically according to your runtime environment, if it couldn't find corresponding Pre-builds, then build script will start automatically, if automatic build doesn't work for you, then please check out Semi-Automatic Build Guide
• If you have any missing node in any open Comfy3D workflow, try simply click Install Missing Custom Nodes in ComfyUI-Manager
• In case there is missing python library, you can check all the python dependencies of my dev environment
• If for some reason your comfy3d can't download pre-trained models automatically, you can always download them manually and put them in to correct folder under Checkpointsdirectory, but please DON'T overwrite any exist .json files
• Docker install please check DOCKER_INSTRUCTIONS.md
• Note: at this moment, you'll still need to install Visual Studio Build Tools for windows and install gcc g++ for Linux in order for InstantNGP & Convert 3DGS to Mesh with NeRF and Marching_Cubes nodes to work, since those two nodes used JIT torch cpp extension that builds in runtime, but I plan to replace those nodes

For manual install

# Fetch newest version of Comfy3D
cd Your ComfyUI Root Directory\ComfyUI\custom_nodes\
git clone https://github.com/MrForExample/ComfyUI-3D-Pack.git
cd ComfyUI-3D-Pack

# Install all dependencies
Your ComfyUI Root Directory\python_embeded\python.exe -s -m pip install -r requirements.txt
Your ComfyUI Root Directory\python_embeded\python.exe install.py

Features:

• For use cases please check out Example Workflows. [Last update: 5/June/2025]

  • Note: you need to put Example Inputs Files & Folders under ComfyUI Root Directory\ComfyUI\input folder before you can run the example workflow

• PartCrafter: wgsxm/PartCrafter

  • Two-model pipeline:
    • Single object mode:
      • Single image → 3D mesh with part segmentation
      • Output:
        • Merged full mesh
        • ZIP archive with individual part meshes
    • Scene mode:
      • Scene image → multiple 3D part meshes
      • Output:
        • Merged full scene mesh
        • ZIP archive with individual object meshes
  • Model weights:
    • PartCrafter (object mode)
    • PartCrafter-Scene (scene mode)
  • Workflows:
    • Single Object
    • Scene

  <video controls autoplay loop src="https://github.com/user-attachments/assets/b80bcc97-7381-4cf7-9ec6-ee48c8d58217"></video> <video controls autoplay loop src="https://github.com/user-attachments/assets/d82f4b32-4916-4286-8478-a86dd5da37a6"></video>

• Hunyuan3D_2.1: Tencent-Hunyuan/Hunyuan3D-2.1

  • Updated two-stage pipeline:
    • Single image → 3D mesh (shape generation)
    • 3D mesh + reference image → 3D mesh with RGB texture
  • Model weights: https://huggingface.co/tencent/Hunyuan3D-2.1
  • Workflows: Full, Shapegen, Texgen

  <video controls autoplay loop src="https://github.com/user-attachments/assets/514bbece-ca19-43cf-83ca-2ccc6c60039e"></video> <video controls autoplay loop src="https://github.com/user-attachments/assets/65f329e7-610f-4520-b0c6-6da62f8209d1"></video> <video controls autoplay loop src="https://github.com/user-attachments/assets/b67a91f4-3d7f-4a2d-9202-c284b013e9fb"></video>

• MV-Adapter: huanngzh/MV-Adapter

  • Two generation methods:
    • IG2MV: Single image + 3D mesh to multi-view images
    • TG2MV: Text prompt + 3D mesh to multi-view images
    • Texturing: Grid image + 3D mesh to textured mesh
  • Model weights: https://huggingface.co/huanngzh/mv-adapter
  • Workflows: IG2MV, T2MV, Texturing

  <video controls autoplay loop src="https://github.com/user-attachments/assets/47b77c9e-a121-45c4-a6d0-f307bda1579c"></video> <video controls autoplay loop src="https://github.com/user-attachments/assets/35c6cee5-d408-4559-88aa-753741eacb95"></video> <video controls autoplay loop src="https://github.com/user-attachments/assets/dcf33c27-0ff7-4379-91b5-a0efe229d63c"></video>

• Stable3DGen: Stable-X/Stable3DGen

  • Two models pipeline:
    • Stable3DGen: Single image to 3D Mesh
    • StableNormal: Image processing for normal map generation
  • Model weights:
    • Stable3DGen: https://huggingface.co/Stable-X/trellis-normal-v0-1
    • StableNormal: https://huggingface.co/Stable-X/yoso-normal-v1-8-1
  • Workflow

  <video controls autoplay loop src="https://github.com/user-attachments/assets/2a38fa0a-4028-4fe5-a67a-bbc7c305da75"></video>

• Hunyuan3D_V2: turbo, mini, fast, multiview

  • Single image to 3D Mesh
  • Multi-views to 3D Mesh with RGB texture
  • Model weights: https://huggingface.co/tencent/Hunyuan3D-2, https://huggingface.co/tencent/Hunyuan3D-2mini
  • Workflows

  <video controls autoplay loop src="https://github.com/user-attachments/assets/ae0f68d8-edd3-4bdd-9a9a-1f5ccc07a3d0"></video>

• TripoSG: VAST-AI-Research/TripoSG

  • Single image (Reference or Scribble) to 3D Mesh
  • Model weights: https://huggingface.co/VAST-AI/TripoSG, https://huggingface.co/VAST-AI/TripoSG-scribble

  <video controls autoplay loop src="https://github.com/user-attachments/assets/7b7969c1-e7c8-465b-9605-8bba7e847dbf"></video>

• TRELLIS: microsoft/TRELLIS

  • Single image to 3D Mesh with RGB texture
  • Model weights: https://huggingface.co/jetx/TRELLIS-image-large

  <video controls autoplay loop src="https://github.com/user-attachments/assets/f569c561-23ea-471f-a9d3-f2e6d1803e00"></video>

• Hunyuan3D_V2 tencent/Hunyuan3D-2

  • Two stages pipeline:
    1. Single image to 3D Mesh shape only
    2. 3D Mesh shape + Single reference image to 3D Mesh with RGB texture
  • Model weights: https://huggingface.co/tencent/Hunyuan3D-2/tree/main
  • Workflows

  <video controls autoplay loop src="https://github.com/user-attachments/assets/da6a8f8b-8dab-4460-90a6-44769ec9bb09"></video>

• Hunyuan3D_V1 tencent/Hunyuan3D-1

  • Two stages pipeline:
    1. Single image to multi-views
    2. Multi-views to 3D Mesh with RGB texture
  • Model weights: https://huggingface.co/tencent/Hunyuan3D-1/tree/main

  <video controls autoplay loop src="https://github.com/user-attachments/assets/2cc31b81-2107-40fe-a531-a783a3f699fc"></video>

• StableFast3D: Stability-AI/stable-fast-3d

  • Single image to 3D Mesh with RGB texture
  • Note: you need to agree to Stability-AI's term of usage before been able to download the model weights, if you downloaded model weights manually, then you need to put it under Checkpoints/StableFast3D, otherwise you can add your huggingface token in Configs/system.conf
  • Model weights: https://huggingface.co/stabilityai/stable-fast-3d/tree/main

  <video controls autoplay loop src="https://github.com/user-attachments/assets/3ed3d1ed-4abe-4959-bd79-4431d19c9d47"></video>

• CharacterGen: zjp-shadow/CharacterGen

  • Single front view of a character with arbitrary pose
  • Can combine with Unique3D workflow for better result
  • Model weights: https://huggingface.co/zjpshadow/CharacterGen/tree/main

  <video controls autoplay loop src="https://github.com/user-attachments/assets/4f0ae0c0-2d29-49f0-a6f2-a636dd4b4dcc"></video>

• Unique3D: AiuniAI/Unique3D

  • Four stages pipeline:
    1. Single image to 4 multi-view